RISE: a fast-readout imager for exoplanet transit timing
I. A. Steele, S. D. Bates, N. Gibson, F. Keenan, J. Meaburn, C. J. Mottram, D. Pollacco, I. Todd
aa r X i v : . [ a s t r o - ph ] S e p RISE: a fast-readout imager for exoplanet transit timing
I. A. Steele a , S. D. Bates a , N. Gibson b , F. Keenan b , J. Meaburn c ,C. J. Mottram a , D. Pollacco b and I. Todd ba Astrophysics Research Institute, Liverpool John Moores University, CH61 4UA, UK; b Astrophysics Research Centre, Queen’s University Belfast, BT7 1NN, UK; c School of Physics and Astronomy, University of Manchester, M13 9PL, UK
ABSTRACT
By the precise timing of the low amplitude (0.005 - 0.02 magnitude) transits of exoplanets around their parentstar it should be possible to infer the presence of other planetary bodies in the system down to Earth-likemasses. We describe the design and construction of RISE, a fast-readout frame transfer camera for the LiverpoolTelescope designed to carry out this experiment. The results of our commissioning tests are described as well asthe data reduction procedure necessary. We present light curves of two objects, showing that the desired timingand photometric accuracy can be obtained providing that autoguiding is used to keep the target on the samedetector pixel for the entire (typically 4 hour) observing run.
Keywords: robotic telescopes, exoplanets, timing, astronomical instrumentation
1. INTRODUCTION
The Liverpool Telescope (LT) is a 2.0 metre fully robotic telescope. The telescope Acquisition & Guidance(A&G) unit can host up to five instruments with the beam able to be directed to one of four side ports via afolding mirror, or, with the folding mirror removed from the beam, to a straight through port. The rapid timeto switch between instruments ( <
30 seconds) combined with automated observing scheduling means that thetelescope is well suited to time variable (on timescales of minutes to years) and rapid reaction astronomy.A key science goal of the Liverpool Telescope is the discovery of Earth mass exoplanets (i.e. planets outsideour own solar system). Two techniques are pursued. The first is via participation in the coordinated followupprogrammes of galactic bulge microlensing events where small anomalies in the light curves act as signaturesof planets orbiting the lens star. An example of the application of this technique in which the LT was involvedwas the recent discovery of a Jupiter/Saturn solar system analogue. The second technique involves the precisetiming of transits of known Jupiter-like exoplanets around their parent star, looking for deviations from theexpected ephemeris due to the influence of other (unseen) planets in the system perturbing the orbit. It is thissecond technique which is the subject of this paper, which describes the construction of a moderately wide field,fast-readout CCD imager for the Liverpool Telescope known as RISE (Rapid Imaging Search for Exoplanets). Inthis paper we will give a brief overview of the science driver for the instrument and the requirements it places onthe design, followed by details of its design and construction. We will also discuss the data reduction procedurefor the instrument, and give the results of our commissioning tests.
2. SCIENCE REQUIREMENTS
The detection of exoplanets is currently of great topical interest in astronomy. As technology and techniquesprogress the main emphasis is naturally moving towards the detection of low mass planets. These objects aredifficult to detect by direct means and it is likely that we will have to wait for an ELT to image the first Earthanalogue system. However, in the mean time, a number of techniques are being used to indirectly detect super-earth massed planets. For an eclipsing system, the idea of using timing residuals to infer the presence of a third,usually lower mass, body is not a new one, going back to the discovery of the outer planets of our own solarsystem. Recent calculations
6, 7 show that a low mass planet moving in a resonance orbit can alter the transittimes of a hot Jupiter planet by > email: [email protected] 1. RISE optical layout. The position of the folding mirror is also indicated. Typical transiting exoplanets so far discovered have magnitudes in the range V ∼ −
12, with amplitudesof 5-20 mmag and durations of 1-3 hours. It can be seen that photometry at the few milli-magnitude level istherefore required. Often signals such as this are dominated by systematic noise sources. To overcome this it isnecessary to image as many comparison stars as possible with similar brightness to the target star. Our field ofview requirement ( ∼ .
3. INSTRUMENT DESIGN AND BUILD3.1 Optical design
The LT has a plate scale of ∼
97 microns/arcsecond, meaning that to cover a 7 . × . ∼ × ∼ . × was therefore required.The overall aim of the optical design (Figure 1) of the field condensing chain was therefore to achieve a7.4 arcmin diameter unvignetted field-of-view with imaging quality better than the CCD pixel size (0.48 arcsec igure 2. Transmission curve for the RISE custom filter constructed from 2mm KG5 + 3mm OG515. diameter) when projected on to the sky. This is achieved with simple off-the-shelf optics which includes a 500 mmfocal length, 63 mm diameter, Melles Griot “Optimized Achromat” as a collimator and a Nikkor 135 mm focallength f/2D commercial compound lens as the camera. A Newport “Precision Achromatic Doublet” field lens,just inside the telescope’s focal plane, places the telescope’s exit pupil on to a filter between the collimator andcamera. This filter is of a similar design to that built for the LT RINGO polarimeter, in this case constructedfrom 2mm Schott KG5 bonded to 3mm Schott OG515 (Figure 2)The transmission surfaces of the two achromatic lenses are coated with three-layer anti-reflection coatingsto minimize losses and ghosts. After the field lens the beam is reflected by a 45 degree plane mirror in anadjustable holder to permit the field-of-view to be centred precisely on the CCD in the camera’s focal planeas well as permitting the exit pupil, after the collimator, to be centred on the optical axis. It was found incommissioning (Section 4) that the optical aims were easily achieved, in fact the useful field-of-view approaches9 arcmin diameter even though vignetted somewhat beyond the design value of 7.4 arcmin. The mechanical design of the instrument (Figures 3 & 4) was driven by the off-the-shelf nature of the opticsused. The collimator lens has a 500mm focal length; this made the instrument too long to be stable without afold mirror. The fold mirror brings the optical axis of the instrument alongside the Acquisition and Guidancebox of the telescope, thereby affording further mechanical brace points. A diagonal bar was fitted spanning thelength of the instrument to eliminate sag through the 90 degree fold. The bottom of the instrument was thenclamped to lower part of the A&G box to eliminate ‘sway’. Due to the physical size of this instrument weightwas going to be a concern, so it was decided to use Aluminum Alloy 5083 H0 for its light weight yet strongtensile strength.There are two lenses in this instrument, a field lens (Newport PAC096) and a collimator lens (Melles GriotLAO 346). The field lens is mounted approximately 10mm inside the focal plane of the telescope in an aluminumblock, using an O ring mounting to allow for the effects of temperature changes at site. The position of thisaluminum block can be adjusted +/- 15mm. Then a fold mirror is mounted 150mm from the telescope focalplane, this mirror has a purpose built kinematic mount and can be adjusted about its centre point easily afterthe entire instrument has been assembled, facilitating alignment. Mounted 350mm further along the optical pathis the collimator lens in a similar mount to the field lens with the same amount of adjustment to allow the focus igure 3. Side view of RISE mechanical design.Figure 4. Angle view of RISE mechanical design.igure 5. Readout time is ∼ ∼ to be set accurately. Finally the Camera (Andor DW435) with bayonet mounted camera lens (Nikkor 135mm)is mounted within a Tufnol plate to electrically insulate it from the rest of the instrument. The Andor CCD camera was connected to an Instrument Control Computer (ICC) via a standard PCI connection.Power for the CCD system is drawn from this connection via a supplemental power connection to the PCI cardfrom the host system. The CCD is constructed with a light insensitive (LISR) 1024x1024 pixel region adjacentto the LSR pixel array of the CCD. This acts as a buffer to hold data being moved from the LSR. This is theframe-transfer aspect of the CCD, and means that when one image acquisition has completed on the LSR, it isimmediately shunted to the LISR. This allows another image to start acquiring very quickly, hence the increasedtime resolution of the camera. This is in contrast to a standard CCD camera, where the image must be read outbefore another is taken. Once an acquisition has started a series of images is acquired by the camera resultingin an exposure-frame transfer-readout cycle (Figure 5.)The software setup for the RISE camera is broken down into two components: the CCD control system andthe CCD acquisition system. The CCD image acquisition system (IAS) is responsible for instructing the CCDhardware to acquire images, and control temperature and onboard acquisition characteristics. This is written inC and makes use of the Andor Linux SDK, allowing camera control software to be written relatively quickly. Thecamera control links the C-level acquisition system to the Java based robotic commands system, which issues theacquisition commands e.g. calibration set (Figure 6). The Java and C implementations are linked by the JavaNative Interface (JNI), allowing the Java based Robotic Control System (RCS) to call the C-based acquisitionroutines.Due to the nature of the image acquisition software the data rate is very high for short exposures. Data canbe acquired in 1x1 or 2x2 binning modes, with minimum exposure cycle times of 1 . . ∼ . ∼ . igure 6. Software command and acknowledgment flow. per hour in 2x2 mode. Another side effect of the high data is the need to run the acquisition process in isolationfrom the the ICS. The usual LT method of acquisition is to take an image and return an acknowledgment to theRCS between each image. However, to do so in this case would create timing delays and introduce an unusuallevel of packet traffic between the IAS and the RCS. The solution to this is to yield control of the instrument tothe IAS and send an acknowledge at the end of a long run.Bias, flat and science frames are taken through two modes of acquisition: twilight calibrate and multrun .Since there is no shutter on the CCD, it can be difficult to acquire a bias image. This must be achieved manuallyby taking zero second exposures with the light path to the CCD obstructed. Dark frames are taken in a similarfashion with the ‘multrun command’. Flats are taken via the ‘twilight calibrate’ command which adapts theexposure time to achieve a similar number of counts on each flat image based on the median count in theprevious image. Finally, the ‘multrun’ command takes a series of images in frame transfer mode. The numberof images required is set along with the exposure time. The acquired data is stored locally on the ICC beforebeing transferred back to LJMU next day, where it is made available via staging on a webserver.
4. COMMISSIONING
RISE was fitted to the Liverpool Telescope in February 2008. Setup was relatively straight forward. Theinstrument was fitted to the telescope without the CCD camera. A test card was clamped to the field lens holderand the holder was adjusted to the focal point of the telescope. A 1 metre focal length telescope (set to infinity)was then positioned where the camera would eventually sit and the collimator lens was adjusted until the testcard could be viewed in focus. The test card was removed and the field lens adjusted away from the focal planeto keep any dust particles out of focus. The mirror was then roughly adjusted by eye to bring all the opticalcomponents concentric within the instrument. The camera was then fitted and images taken of the inside of theenclosure. A combination of small adjustments of the telescope fold mirror in the A&G box and the fold mirrorwithin the instrument itself were then made to centre the image.On sky commissioning took place over the period 20-23rd February 2008. The field of view (with somevignetting in the corners) was found to be 9.4 x 9.4 arcminutes, corresponding to a pixel scale of 0.55 arcsec/pixel(Figure 7). No significant variation in point spread function across the field could be detected. The detectorsystem read-noise was measured at 10 electrons rms, and the gain as 2.4 electrons/count. Linearity testingshowed good performance up to the ADC conversion limit to 65,000 counts with 1x1 binning and around 40,000counts with 2x2 binning. With the short exposure times used for the instrument no significant dark currentcould be detected.
5. DATA REDUCTION
The reduction of data to milli-mag accuracies is a complex task, and requires a good understanding of thesystematic errors. At present our data reduction procedures are still at an experimental stage, although we canalready demonstrate high quality photometry and timing results (Figure 8). igure 7. A comparison of raw images from the RISE (left hand side) and RATCam (right hand side) cameras of theLiverpool Telescope to the same scale.
The initial stages the raw images are first put through a python/pyraf script that updates the headers that areneeded at the reduction or lightcurve fitting stage, including the heliocentric Julian date and airmass (calculatedfrom DATE-OBS, RA and DEC), plus other instrument characteristics such as the gain, read-noise and filter.Due to the instrument not having a shutter we try both bias images and bias strips to debias the images. Flatfielding is also tricky. We currently do not understand how the flat field structure changes during the course of anight (due to scattered light in the telescope and optics) and therefore we prefer not to flat field in cases whereautoguiding is stable. This is because the ratio of the stars fluxes on the chip will stay constant and flat fieldingeffects will be removed to first order during sky subtraction anyway. If however there is some drift (e.g. dueto lack of an appropriate guide star) flat fielding must be used to remove some (but not all) of the systematiceffects. An example of this is shown in Figure 9 where autoguiding was lost half way through the run (Figure10) - the detrimental effect on the light-curve is obvious.Aperture photometry is performed on the images using IRAF/daophot via a python/pyraf script that uses alist of apertures to calculate the relative flux of the target star (given a list of reference stars coordinates) and therespective errors. Typically an aperture of 10 pixels is used (in 2x2 binning mode). Following this the lightcurvesare then normalized using an airmass (or time) function and a model is fitted to the lightcurves according toan analytic formula using a Monte Carlo Markov Chain code with suitable priors on known stellar and orbitalparameters. This allows us to calculate the radius, inclination and central transit times. It is the variation ofthese central transit times from the predicted ephemeris that will provide evidence of a third body.
6. CONCLUDING REMARKS
RISE was a fast-track instrument for the Liverpool Telescope, which was designed, built and commissioned within1 year of the initial science requirement being identified. Although work is still underway to better understandthe effect of scattered light on the observations, it is already clear that the timing precision obtainable is sufficientto carry out our programme of transit followup, which has therefore begun routine data taking. In addition we R e l a t i v e F l u x JDnorm_TRES3_20080308.flux F i g u r e . L i g h t c u r v e a nd m o d e l r e s i du a l s f o r t h e t r a n s i t i n g e x o p l a n e t T R E S . r e l f l u x GJ436_20080422 F i g u r e . R I S E li g h t c u r v e o f G J . N o t e h o w t h e l o ss o f t h e a u t og u i d e r h a l f w a y t h r o u g h t h e s e q u e n c e o f e x p o s u r e s a nd t h e r e s u l t i n g d r i f t o f t h e t a r g e t a c r o ss t h e CC D c a u s e s p r o b l e m s w i t h t h e ph o t o m e t r y . B e f o r e l o ss o f t h e a u t og u i d e r h o w e v e r , t h ee c li p s e ( w i t h a d e p t h o f . m ag n i t ud e s ) i s e a s il yv i s i b l e . y xTRES3_20080308 target positions 220 230 240 250 260 270 280 200 210 220 230 240 250 260 270 y xGJ436_20080422 target positions F i g u r e . T a r g e t ss t a r c e n t r o i d s f o r t h e o b s e r v a t i o n s o f T R E S ( a u t og u i d e d ) a nd G L ( n o t a u t og u i d e d ) . re pleased to note that it is also attracting interest from other science users of the telescope attracted by itslow overheads for timing experiments and its relatively wide field. ACKNOWLEDGMENTS
We thank the staff of the QUB Physics workshop for their excellent work in constructing RISE. The LiverpoolTelescope is operated on the island of La Palma by Liverpool John Moores University in the Spanish Observatoriodel Roque de los Muchachos of the Instituto de Astrofisica de Canaries with financial support from the UK Scienceand Technologies Facilities Council.
REFERENCES [1] Steele, I. A., Smith, R. J., Rees, P. C., Baker, I. P., Bates, S. D., Bode, M. F., Bowman, M. K., Carter,D., Etherton, J., Ford, M. J., Fraser, S. N., Gomboc, A., Lett, R. D. J., Mansfield, A. G., Marchant, J. M.,Medrano-Cerda, G. A., Mottram, C. J., Raback, D., Scott, A. B., Tomlinson, M. D., and Zamanov, R.,“The Liverpool Telescope: performance and first results,” in [
Proceedings of the SPIE, Volume 5489, pp.679-692 (2004). ], Oschmann, J. M., ed., 679–692 (Oct. 2004).[2] Mottram, C. J., Steele, I. A., and Morales, L., “Design of low cost and reliable instrumentation for robotictelescopes,” in [
Ground-based Instrumentation for Astronomy. Edited by Alan F. M. Moorwood and IyeMasanori. Proceedings of the SPIE, Volume 5492, pp. 677-688 (2004). ], Moorwood, A. F. M. and Iye, M.,eds., 677–688 (Sept. 2004).[3] Fraser, S. and Steele, I. A., “Robotic telescope scheduling: the Liverpool Telescope experience,” in [
Optimiz-ing Scientific Return for Astronomy through Information Technologies. Edited by Quinn, Peter J.; Bridger,Alan. Proceedings of the SPIE, Volume 5493, pp. 331-340 (2004). ], Quinn, P. J. and Bridger, A., eds.,
Pre-sented at the Society of Photo-Optical Instrumentation Engineers (SPIE) Conference , 331–340 (Sept.2004).[4] Burgdorf, M. J., Bramich, D. M., Dominik, M., Bode, M. F., Horne, K. D., Steele, I. A., Rattenbury, N.,and Tsapras, Y., “Exoplanet detection via microlensing with RoboNet-1.0,”
PLANSS , 582–588 (Apr.2007).[5] Gaudi, B. S., Bennett, D. P., Udalski, A., Gould, A., Christie, G. W., Maoz, D., Dong, S., McCormick,J., Szyma´nski, M. K., Tristram, P. J., Nikolaev, S., Paczy´nski, B., Kubiak, M., Pietrzy´nski, G., Soszy´nski,I., Szewczyk, O., Ulaczyk, K., Wyrzykowski, L., DePoy, D. L., Han, C., Kaspi, S., Lee, C.-U., Mallia, F.,Natusch, T., Pogge, R. W., Park, B.-G., Abe, F., Bond, I. A., Botzler, C. S., Fukui, A., Hearnshaw, J. B.,Itow, Y., Kamiya, K., Korpela, A. V., Kilmartin, P. M., Lin, W., Masuda, K., Matsubara, Y., Motomura,M., Muraki, Y., Nakamura, S., Okumura, T., Ohnishi, K., Rattenbury, N. J., Sako, T., Saito, T., Sato,S., Skuljan, L., Sullivan, D. J., Sumi, T., Sweatman, W. L., Yock, P. C. M., Albrow, M. D., Allan, A.,Beaulieu, J.-P., Burgdorf, M. J., Cook, K. H., Coutures, C., Dominik, M., Dieters, S., Fouqu´e, P., Greenhill,J., Horne, K., Steele, I., Tsapras, Y., Chaboyer, B., Crocker, A., Frank, S., and Macintosh, B., “Discoveryof a Jupiter/Saturn Analog with Gravitational Microlensing,” Science , 927– (2008).[6] Holman, M. J. and Murray, N. W., “The Use of Transit Timing to Detect Terrestrial-Mass ExtrasolarPlanets,”
Science , 1288–1291 (Feb. 2005).[7] Agol, E., Steffen, J., Sari, R., and Clarkson, W., “On detecting terrestrial planets with timing of giantplanet transits,”
MNRAS , 567–579 (May 2005).[8] Cameron, A. C., Bouchy, F., H´ebrard, G., Maxted, P., Pollacco, D., Pont, F., Skillen, I., Smalley, B., Street,R. A., West, R. G., Wilson, D. M., Aigrain, S., Christian, D. J., Clarkson, W. I., Enoch, B., Evans, A.,Fitzsimmons, A., Fleenor, M., Gillon, M., Haswell, C. A., Hebb, L., Hellier, C., Hodgkin, S. T., Horne, K.,Irwin, J., Kane, S. R., Keenan, F. P., Loeillet, B., Lister, T. A., Mayor, M., Moutou, C., Norton, A. J.,Osborne, J., Parley, N., Queloz, D., Ryans, R., Triaud, A. H. M. J., Udry, S., and Wheatley, P. J., “WASP-1b and WASP-2b: two new transiting exoplanets detected with SuperWASP and SOPHIE,”
MNRAS ,951–957 (Mar. 2007).9] Pollacco, D., Skillen, I., Collier Cameron, A., Loeillet, B., Stempels, H. C., Bouchy, F., Gibson, N. P., Hebb,L., H´ebrard, G., Joshi, Y. C., McDonald, I., Smalley, B., Smith, A. M. S., Street, R. A., Udry, S., West,R. G., Wilson, D. M., Wheatley, P. J., Aigrain, S., Alsubai, K., Benn, C. R., Bruce, V. A., Christian, D. J.,Clarkson, W. I., Enoch, B., Evans, A., Fitzsimmons, A., Haswell, C. A., Hellier, C., Hickey, S., Hodgkin,S. T., Horne, K., Hrudkov´a, M., Irwin, J., Kane, S. R., Keenan, F. P., Lister, T. A., Maxted, P., Mayor,M., Moutou, C., Norton, A. J., Osborne, J. P., Parley, N., Pont, F., Queloz, D., Ryans, R., and Simpson,E., “WASP-3b: a strongly irradiated transiting gas-giant planet,”
MNRAS , 1576–1584 (Apr. 2008).[10] Wilson, D. M., Gillon, M., Hellier, C., Maxted, P. F. L., Pepe, F., Queloz, D., Anderson, D. R., CollierCameron, A., Smalley, B., Lister, T. A., Bentley, S. J., Blecha, A., Christian, D. J., Enoch, B., Haswell,C. A., Hebb, L., Horne, K., Irwin, J., Joshi, Y. C., Kane, S. R., Marmier, M., Mayor, M., Parley, N., Pollacco,D., Pont, F., Ryans, R., Segransan, D., Skillen, I., Street, R. A., Udry, S., West, R. G., and Wheatley, P. J.,“WASP-4b: A 12th Magnitude Transiting Hot Jupiter in the Southern Hemisphere,”
ApJ , L113–L116(Mar. 2008).[11] Anderson, D. R., Gillon, M., Hellier, C., Maxted, P. F. L., Pepe, F., Queloz, D., Wilson, D. M., CollierCameron, A., Smalley, B., Lister, T. A., Bentley, S. J., Blecha, A., Christian, D. J., Enoch, B., Hebb, L.,Horne, K., Irwin, J., Joshi, Y. C., Kane, S. R., Marmier, M., Mayor, M., Parley, N. R., Pollacco, D. L.,Pont, F., Ryans, R., S´egransan, D., Skillen, I., Street, R. A., Udry, S., West, R. G., and Wheatley, P. J.,“WASP-5b: a dense, very hot Jupiter transiting a 12th-mag Southern-hemisphere star,”
MNRAS , L49+(Apr. 2008).[12] Steele, I. A., Bates, S. D., Carter, D., Clarke, D., Gomboc, A., Guidorzi, C., Melandri, A., Monfardini,A., Mottram, C. J., Mundell, C. G., Scott, A. B., Smith, R. J., and Swindlehurst, J., “RINGO: a novelring polarimeter for rapid GRB followup,” in [
Ground-based and Airborne Instrumentation for Astronomy.Edited by McLean, Ian S.; Iye, Masanori. Proceedings of the SPIE, Volume 6269, pp. 62695M (2006). ], Presented at the Society of Photo-Optical Instrumentation Engineers (SPIE) Conference (July 2006).[13] Mandel, K. and Agol, E., “Analytic Light Curves for Planetary Transit Searches,”
ApJ580