Clusters And Variability

Group Members

Our team currently consists of:

• Dr. Ray Butler (lecturer in Physics & Astronomy, and CfA secretary)
• Mr. Brendan Sheehan (PhD student & IRCSET scholar)
• Mr. Leon Harding (PhD student) 
• (…and best of luck to our former member Mr. John Chambers, who has returned to NASA Goddard after completing his MSc thesis here, as part of the European Masters in Astronomy)

- plus numerous collaborators within NUI Galway and elsewhere.

As the title of this page indicates, we have fingers in many different pies. The common thread in all our endeavours is the development and use of improved astronomical/imaging technologies, principally in the software domain, to attain a variety of scientific goals. We therefore have considerable experience in:

• The end-to-end process of acquiring (or simulating) and processing astronomical images, including:
  • Exposure Time Calculator (ETC) development, to model instrument throughput as a function of wavelength
  • Point-Spread-Function (PSF) modelling, including time-variable seeing distributions and spatially-variable PSFs (SV-PSFs)
  • Simulating images from various instruments, incorporating high levels of realism in both astrophysical and instrumental inputs
  • Design of custom narrowband optical filters, for spectrophotometry
  • Deriving schemes for attaining the maximum signal-to-noise ratio in imaging, taking observational duty cycle into account
  • Developing automated data sorting and reduction "pipeline" software, with sophisticated exception handling
• Advanced image processing/analysis techniques, such as:
  • Deconvolution, with SV-PSFs and subsampling
  • Crowded-field photometry – both PSF-fitting (DAOPHOT) and image subtraction (ISIS)
  • Time-series photometric analysis
  • Astrometric calibration
  • Post-exposure imaging sharpening (PEIS) and frame selection (“Lucky

Our research interest in these technologies drives our scientific research programmes – and vice-versa. It is a symbiotic relationship. We are constantly posing questions like “how can we make better observations of this type of target?” and “how can we extract more information from this type of data?”

If you are a researcher and find yourself asking the same kind of question, then we’d like to hear from you. Perhaps we can help you – and perhaps you can help us.

Already, our skills base has led to much collaborative work in different fields, in particular assisting with and participating in the observational programmes on ultra-cool dwarfs, and pulsars/supernova remnants, which are led by our colleagues Aaron Golden and Andy Shearer in the CfA.

Star Clusters

We are interested in the stellar populations of star clusters, primarily the globular clusters of the Milky Way galaxy. The rich ecology of populations that we have studied in such clusters includes several classes of variable stars, close binary stars, pulsars, and even an extra-solar planet. At present, we are observing clusters with our custom-designed filters, to measure the variation in absorption of starlight due to varying abundances of carbon and/or nitrogen. The presence of these variations from star to star within some clusters is at odds with the standard model of all stars having been formed at the same time, from the same mix of chemical elements (i.e. the same "metallicity"). Our results to date indicate that these variations are indeed primordial in origin. Kieran Forde is currently on an extended placement at the University of California at Santa Cruz/Lick Observatory, investigating the extension of this field to the regime of extra-galactic globular clusters.

New work underway in this area includes investigation of the use of deconvolution of early-epoch HST data, for more accurate astrometry in cluster cores. This will be used to extend back the baseline in the search for high-proper motion stars, which could be evidence of an intermediate-mass black hole.

Variability

Our interest in variable stars began in the 1990s, with the development at Galway of the TRIFFID high angular resolution camera. TRIFFID used imaging photon-counting detectors (MAMAs) for very high-speed imaging, and PEIS post-processing, to achieve what were then some of the sharpest long-exposure ground-based optical images ever taken. This instrument was put to use to target low-mass X-ray binaries, buried in the highly crowded, scarcely resolvable cores of globular clusters. A bonus of this work was the detection or confirmation in the cluster centres of a large number of hitherto unknown variable stars, of various types (RR Lyrae, Type 2 Cepheid, possible eclipsing). Upper limits were also set on the occurrence of dwarf novae.

Subsequently, the need for instruments like TRIFFID to obtain the highest angular resolution for this kind of variability work was superceded by the development of effective image-subtraction algorithms, like ISIS. Although we used ISIS on our later TRIFFID data and the enhanced angular resolution did help, it was clear that very high photometric signal-to-noise (the “depth” of the dataset) made up most or all of the ground lost when resolution was poorer. This handed the initiative back to regular CCD cameras, with their much higher quantum efficiency (q.e.), and much simpler calibration, than TRIFFID-type imaging photon-counting detectors.

But now there’s a new game in town. The development of a new CCD camera technology, dubbed the Low Light Level (L3) CCD, can give us the best of both worlds: the high q.e. & ease of calibration of a CCD, the high speed & lack of readout dead-time of an imaging photon-counting detector, and negligible detector noise. With our colleagues in the Applied Imaging group and Applied Optics Centre, we have built a new camera system based on L3-CCD technology, which we have called GUFI - the Galway Ultra Fast Imager.  This camera is suitable for PEIS & Lucky Imaging work, as well as optimised longer-exposure time-series photometry with a 100% duty cycle. The system received "first light" in 2006 on the 1.5m telescope at Loiano, Italy. Brendan Sheehan has been working on building and calibrating the L3-CCD setup - including simulations of its performance, and determining how to optimize its use for a given observational scenario. Brendan has also written a very efficient and highly functional automated pipeline for L3-CCD data reduction and differential photometry. John Chambers designed a two-beam "outrigger" variant for the GUFI optics; the purpose of the steerable second beam would be to help to locate suitably bright comparison stars for diffferential photometry in sparse fields, even when such stars lay well outside the small field of view of the main beam. Leon Harding is currently converting the old (HP-UX/Digital Unix) PEIS code for the MAMA photon-stream data format, to a new version that will work with FITS image datacubes from the L3-CCD camera.

GUFI's first target was another type of (potentially) variable object that interests us – brown dwarfs/ultra-cool dwarfs. Aaron Golden’s group have been highly successful in radio studies of these odd little objects, and we meld with this group to provide optical/IR timeseries photometry. This has been Caoilfhionn Lane’s main PhD activity.

Still in the domain of variability, we are also interested in detecting extra-solar planets, aka exo-planets, and measuring their properties. This interest originally arose out of the anticipated large numbers of such planets in globular clusters, which however subsequently failed to materialize despite lengthy observational campaigns by a number of groups. It is now (since 2005) known that in fact the metallicity of these clusters is too low for "normal" planets to have formed, so we have adapted our efforts to conduct targeted observations of known, suspected, or possible transiting exoplanets around more nearby stars. If a planet orbits its parent star in the same plane as our line of sight to the star, it passes (transits) across the face of the star and thereby blocks a tiny proportion (~2% or generally much less) of the star's light. This is detectable as a dip in the brightness of the star, measured over the course of a series of repeated images. This observing procedure - transit photometry - is limited by both atmospheric scintillation and inefficiencies or noise imposed by detector readout. Our approach is to minimise the latter limitation, by the development of a new camera system based on L3-CCD technology. The system received "first light" in 2006 on the 1.5m telescope at Loiano, Italy. We attempted to make other observations with a regular CCD on the Faulkes 2m telescope at Maui, Hawaii, in conjunction with transition year students at local schools, but were stymied by tropical typhoons or schedulign overrides every time.

Joining the Group

If you are interested in joining the group to carry out an MSc, PhD or as a postdoc, then please contact Ray Butler in the first instance.
Postgrads: We will be supporting a small number of applications for IRCSET and NUI Galway College of Science PhD fellowships, commencing September/October 2008.
Postdocs: We would be delighted to support postdoctoral applications under the FP7 Marie Curie and IRCSET programmes. We also hope to be successful with our FP7 ITN proposal; watch this space…

 
One last thing: you don’t have to speak IRAF/Pyraf to join our group – but you will be fluent by the time you leave!

• Gregg Hallinan 
• Aaron Golden

Collaborators

• Gerry Doyle and Tony Antonova ( Armagh Observatory, N. Ireland)
• Walter Brisken ( National Radio Astronomy Observatory, New Mexico)
• Bob Zavala and Fred Vrba ( United States Naval Observatory, Flagstaff, Arizona)
• Richard Boyle ( Vatican Observatory Research Group, Steward Observatory, University of Arizona)

The Discovery of Radio Pulsations from Ultracool Dwarfs

Brown dwarfs occupy the mass gap between planets and stars and are thought to be one of the most populous objects in our Galaxy. They have a mass below that necessary to maintain hydrogen-burning nuclear fusion reactions in their cores and are therefore much cooler and dimmer than main sequence stars. This makes them very difficult to detect, and although astronomers have known of their existence for decades it wasn't until 1995 that a brown dwarf was finally found.

Figure 1: Artist's impression of the "super-aurorae" present at the magnetic poles of these radio emitting dwarfs which are responsible for the radio pulsations.

In recent years it has been discovered that these brown dwarfs can be extremely bright sources of radio emission. Up to now it has been unclear how these failed stars can produce such high levels of this nature of radiation. Initially, it was assumed that it was the same kind of radio emission as that detected from stars such as our Sun. For such stars, the radio emission is produced by high energy electrons in the star's corona which are trapped spiralling in the star's magnetic field.

Figure 2: Two images of the location of the binary L dwarf 2MASSW J0746425+2000321 taken with the VLA at a frequency of 4.88 GHz . The position of the binary brown dwarf is marked by a white arrow in both cases. A field source is located 10 arcseconds from the brown dwarf and is present in both images to the upper left hand side of the target. The left hand image was taken during the interpulse phase, when the binary brown dwarf is too faint to be detected. The right hand image was taken during one of the pulses with the brown dwarf now outshining the nearby field source.

However, our recent observations conducted with the Very Large Array radio telescope in New Mexico, together with optical telescopes at the US Naval Observatory and Vatican Observatory, have shown that this model is incorrect. We have detected extremely bright periodic pulses of radiation from a number of these objects which cannot be explained by the conventional processes associated with stellar radio emission. During these pulses, these supposedly failed stars are tens of thousands of times brighter than our Sun at radio frequencies! Instead a much more exotic process is required to explain such bright radio emission.

It turns out that the answer to this mystery is not to be found in the study of the radio emission from the stars but instead from the planets in our Solar System. All the magnetized planets, including Earth, are observed to emit extremely bright radio emission from their magnetic polar regions. Indeed, Jupiter can produce radio emission at low frequencies brighter than that detected from the Sun. This radiation is not produced by the same mechanism responsible for stellar radio emission but rather by a coherent process, the electron cyclotron maser, that can amplify the radiation to extremely high levels. In the Earth's case, the maser radiation is produced when the Solar wind slams into the planet's magnetosphere accelerating electrons into the polar regions. These electrons produce extremely bright radio emission and subsequently impact the Earth's ionosphere to produce aurorae visible from the Earth's surface.

Movie1: Animated gif of the radio emission from the M9 dwarf TVLM 513-46546 detected with the VLA at 8.44 GHz . The time between each bright pulse corresponds to 1.958 hours, which is the period of rotation of the dwarf. The movie was created from the data obtained with the VLA using the ParselTongue interface to the AIPS astronomical data analysis package.

A very similar process is believed to apply to brown dwarfs, albeit producing radio emission many orders of magnitude brighter than that detected from the planets. The resulting radiation, which is very strongly beamed perpendicular to the magnetic field of the brown dwarf, sweeps Earth once per rotation period of the dwarf to produce the bright pulses. However, it remains a mystery how the high energy electrons which produce the radio emission are continuously accelerated into the magnetic poles of the dwarf. What has been established is that this radio emission requires these brown dwarfs to possess very powerful, large-scale magnetic fields as strong as those detected from the most magnetically active main sequence stars.

Figure 3: Time series of the radio emission detected with the VLA from the M9 dwarf TVLM 513-46546. Every 1.958 hours a periodic pulse is detected when extremely bright, beams of radiation originating at the poles sweep Earth when the dwarf rotates.

The periodic pulses detected from brown dwarfs are very similar in nature to those observed from one of the most exotic classes of object in our Universe, pulsars. Pulsars are produced during supernovae, when a massive star explodes and it's core collapses into a rapidly spinning neutron stars. Beams of radiation are emitted from the magnetic poles of the neutron star and, as is the case for brown dwarfs, sweep Earth with rotation of the star. How pulsars produce this radiation has been one of the great puzzles in astrophysics for nearly 40 years. The difficulty in establishing this emission mechanism is grounded in our lack of understanding of the behaviour of plasma in the extreme conditions associated with pulsars. Brown dwarfs are now the second class of stellar object known to produce persistent levels of extremely bright, coherent radiation. Moreover, this emission is also manifested in the detection of periodic pulses. However, in the case of brown dwarfs, both the source conditions and the emission mechanism are reasonably well understood. It is hoped that the study of these brown dwarfs may provide vital clues to unlocking the long-standing puzzle of how pulsars produce their radio emission.

Movie2: Same as Movie 1, with a much wider field of view. A field source can be seen to emit fairly steadily in the upper right hand corner of the animated gif while TVLM 513-46546 is located in the lower left hand corner of the animated gif and is seen to periodically pulse. The time between each bright pulse corresponds to 1.958 hours, which is the period of rotation of the dwarf. The movie was created from the data obtained with the VLA using the ParselTongue interface to the AIPS astronomical data analysis package.

"This work is supported by Science Foundation Ireland under its Research Frontiers Programme, and the Irish Research Council for Science, Engineering and Technology (IRCSET)."

• Disclaimer
• Privacy
• Copyright
• Accessibility