RATS-Kepler -- a deep high cadence survey of the Kepler field
We outline the purpose, strategy and first results of a deep, high cadence, photometric survey of the Kepler field using the Isaac Newton Telescope on La Palma and the MDM 1.3m Telescope on Kitt Peak. Our goal was to identify sources located in the Kepler field of view which are variable on a timescale of a few mins to 1 hour. The astrophysically most interesting sources would then have been candidates for observation using Kepler using 1 min sampling. Our survey covered 42% of the Kepler field of view and we have obtained light curves for 7.1 objects in the range 13. We have discovered more than 100 variable sources which have passed our two stage identification process. As a service to the wider community, we make our data products and cleaned CCD images available to download. We obtained Kepler data of 18 sources which we found to be variable using our survey and we give an overview of the currently available data here. These sources include a pulsating DA white dwarf, eleven Sct stars which have dominant pulsation periods in the range 24 min to 2.35 hrs, three contact binaries, and a cataclysmic variable (V363 Lyr). One of the Sct stars is in a contact binary.
keywords:Astronomical data bases: surveys; Physical data and processes: asteroseismology; stars: variable - white dwarf - Scuti
The prime objective of the Kepler mission is to detect Earth sized planets orbiting Solar type stars in the habitable zone (Koch et al 2010). It does this by detecting transits of the host star by the orbiting exoplanet. The lightcurves which Kepler obtained extended over many months and have a precision of parts per million. These data allow models of stellar structure to be tested in a way that has not been possible before (e.g. Bedding et al. 2011). Furthermore, it has led to the unexpected discovery of extreme binary systems such as the ’Heartbeat’ stars which are excellent tests of binary and stellar models (Welsh et al. 2011, Thompson et al. 2012).
Asteroseismology provides the means to probe the masses and compositions of stellar interiors; determine stellar internal rotation profiles; the extent of instability strips and therefore test models of stellar structure and evolution (e.g. Chaplin et al. 2011). To study compact objects such as pulsating white dwarfs, relatively high cadence observations are essential. For the vast majority of observations made using Kepler, the effective exposure time is 30 mins (’Long Cadence’). However, for a much more limited number of stars (512) a shorter effective exposure of 1 min is possible (’Short Cadence’).
Before the launch of Kepler, an extensive programme to identify bright G/K dwarfs with minimial stellar activity was carried out. Although a small number of photometric variability surveys were carried out pre-launch (e.g. Hartman et al. 2004, Pigulski et al. 2009, Feldmeier et al 2011) they were either not especially deep, did not have wide sky coverage or did not have a cadence shorter than a few minutes. To fill this gap we started a photometric variability survey (RATS-Kepler) in the summer of 2011 using the Wide Field Camera on the Isaac Newton Telescope (INT). Sources which were considered astrophysically interesting based on their light curve and colour would then have been the subject of bids to obtain Kepler Short Cadence observations.
2 Photometric Observations
Our strategy is a modified version of that used by us in the RApid Temporal Survey (RATS) which was carried out using the INT between 2003 and 2010 (Ramsay & Hakala 2005, Barclay et al. 2011). In that project we obtained a series of 30 sec exposures of a given field in white light for 2.0–2.5 hrs. The resulting lightcurves had a resulting cadence of 1 min and, for sources brighter than =21, the standard deviation () of the light curves was 0.024 mag (Barclay et al. 2011). It led to the discovery of a rare double-mode pulsating sdB star (Ramsay et al. 2006, Baran et al. 2011), pulsating white dwarfs and several dozen distant Sct or SX Phe stars (Ramsay et al. 2011).
Since the Kepler field of view (116 square degrees) is more than twice
the area covered by the RATS project, we decided to increase the
number of fields observed per night by obtaining a one hour (rather
than a two hour) sequence of short exposures per pointing. Since the
photometric precision of Kepler Short Cadence observations reduces
from 12.9 percent at =19 to 32.4 percent at =20 (this compares
with 0.85 percent at
During the summer of 2011 and 2012 we obtained data using the 2.5m INT located on the island of La Palma, and the 1.3m MDM McGraw-Hill Telescope located on Kitt Peak (see Table 1 for details). Our observations cover 42 percent of the Kepler field. In Figure 1 we show the position of stars observed in our survey in equatorial coordinates.
2.1 Isaac Newton Newton Telescope
The INT Wide Field Camera has 4 CCDs and covers 0.29 degrees squared. The deadtime was 30 sec, giving a cadence of 50 sec. We are sensitive to flux variations on timescales as short as a few mins in sources with a magnitude in the range 13.521. Tables A1 and A2 show the dates and field centers of each pointing.
2.2 MDM Telescope
The red4k detector was used for two nights and the MDM4k detector for
five nights on the MDM
|11–17 Jul 2011||INT||49|
|01–10 Aug 2011||INT||58|
|16–22 May 2012||MDM||26|
|03–12 Aug 2012||INT||55|
2.3 Image Reduction
The data were corrected for the bias level and were flat-fielded using
3 Data Analysis
We broadly follow the same data reduction and analysis strategy as we used for the RATS project, which is described in Barclay et al. (2011). However, we now outline some features specific to the RATS-Kepler project.
3.1 Extracting light curves
The Kepler field extends 6–21 degrees above the Galactic plane. Each of our individual fields are therefore relatively crowded at low latitudes or surprisingly sparse at higher latitudes. For sparse fields we used sextractor (Bertin & Arnouts 1996) to extract magnitudes using aperture photometry. Differential magnitudes were determined by comparing the magnitude of each star with the mean brightness of the 3–10th most brightest stars in the image (the results were very similar if we chose, say, the 4–20th most brightest stars). For more crowded fields we used diapl2, an updated version of diapl (Wozniak 2000), which extracts photometry by applying the well established ‘Difference Imaging Subtraction’ method (Alard & Lupton 1998).
Despite the fact that differential photometry has been performed we find that the photometry of certain fields suffer from systematic trends in the data – ie the light curves derived from the same detector can show similar features. This effect can be seen in many other large scale surveys, including Kepler (Kinemuchi et al 2012). In our case, the systematic trends will largely result from the fact that for good technical reasons we do not use the autoguider for our INT observations. To ensure that stars remain roughly at the same position on the detector we apply manual corrections to the pointing.
However, we aimed to remove the effects of systematic trends by appling the SYSREM algorithm (Tamuz et al. 2005) to the light curves derived from each CCD individually using a varying number of cycles as described in Tamuz (2006) (see also Barclay et al. 2011). For a small number of fields it was not possible to detrend the data (mainly because of the low number of stars available). Using a faint limit of and a bright limit of , we have obtained a total of 7.1 detrended light curves.
3.2 Identifying variable candidates
Identifying bona fide variable stars from a large sample of light curves is not a trivial task. Here we use a two stage process of identifying variable stars. In the first stage we use different statistical tests to obtain a sample of candidate variables. In the second stage we manually perform a quality assessement of each light curve and associated images to remove sources which have been spuriously identified as variable.
Different statistical tests are better suited to identifying different kinds of variability. For instance, the Lomb Scargle (LS) Periodogram (Lomb 1976, Scargle 1982) is particularly well suited to detecting pulsating variables where the pulsation period is shorter than the duration of the light curve. On the other hand, the Alarm test (Tamuz, Mazeh & North 2006) and the Analysis of Variance (AoV) test (Schwarzenberg-Czerny 1989, 1996 and Devor 2005 for the implementation used by VARTOOLS) are suitable for identifying eclipsing or contact binaries, whilst the test (where the model is the mean magnitude of the light curve) is good for detecting flare stars (see Graham et al. 2013 for a recent review of which tests are best suited to identifying specific kinds of variable star).
The VARTOOLS suite of software (Hartman et al. 2008) allows the parameterisation of large numbers of light curves using many different statistical tests in a quick and simple manner. We apply the following tests on each of our light curves: the LS Periodogram; the Alarm test and the AoV test. We also determine the value and standard deviation () for each light curve after applying a 5 clipping to each light curve. A file containing the positions and colours of all sources along with their photometric variability parameters can be downloaded via Armagh Observatory Web Site (star.arm.ac.uk/rats-kepler). Table 7 outlines the full set of parameters which are given in this FITS file.
As an example of how different tests compare, we show in Figure 2 the results of the , LS and AoV tests on a field which contains KIC 3223460 which has a dominant pulsation period of 24.2 mins (see Table 2). Plotting the of each light curve as a function of mag shows that KIC 3223460 has a greater than the main distribution of sources with similar magnitude. However, there is (naturally) no information on the timescale of variability. On the other hand, the LS and AoV test clearly identify KIC 3223460 as being strongly variable on a period of 24 mins.
The main goal of our survey is to identify compact pulsating stars in the Kepler field which would then have been the subject of bids to observe them in Short Cadence mode using Kepler. As demonstrated in Figure 2, the LS Periodogram is efficient at identifying these sources in our data. The LS Periodiogram as implemented in VARTOOLS determines the frequency of the highest peak in the power spectrum (which we define as the ‘Period’ of variability even if the source cannot be verified as stricty periodic) and the False Alarm Probability of this peak being statistically significant. For each light curve we obtained an LS power spectrum and performed the AoV test in the frequency interval corresponding to the Nyquist frequency (847.1 cycles/day – which equates to a period of 1.7 mins – for the INT data and 600 cycles/day – which equates to a period of 2.4 mins – for the MDM data) and 21.49 cycles/day – which equates to a period of 67 mins – (which is the mean duration of the INT light curves).
For the purposes of selecting an initial sample of candidate variables, we use the LS test. In the absence of red noise and systematic trends, a peak in the power spectrum with log FAP=–2.5 is likely to be significantly variable at the 3 confidence level. However, since the seeing and sky brightness can vary from field to field, and the success of the detrending alogrithim can vary from chip to chip, the threshold for identifying variables can be more negative than log FAP = –2.5.
We determined the median value of the log FAP statistic for the light curves in each field-chip combination. We then ordered our sources by this median log FAP and made seven sub-sets containing 10 stars each (the remaining 8787 stars which were in field-chip combinations with the highest median log FAP were discarded). For the subset with the least negative mean value for log FAP (Figure 3), there were 911 sources which has a log FAP -2.5 (or 0.91 percent of sources in the sub-set).
Rather than using a fixed threshold for the log FAP to provide an initial selection of candidate variables, we used the Median Absolute Deviation (MAD) to provide a means of identifying sources which were ’outliers’ in the Period - log FAP distribution. The MAD is defined for a batch of parameters as
We ordered the data by Period and then into 2 min time bin intervals and derived the MAD for each bin. Candidate variables are then selected so that variable sources obey (log FAP) MADn + Median, where is an integer which defines how far a source is from the local median log FAP. It is selected largely by trial and error – too high a value of will select only the most strongly variable sources, but too low a value of will produce large quantities of candidate variables, all of which require manual verification. (To be selected as a candidate variable, a source also has to have log FAP -2.5). The selection of variables using the MAD statistic was done on each subset of 10 stars separately then combined according to the value that was used. We found that for =18, 227 stars (or 0.032 percent of the total) were selected as candidate variables (=16 368 stars, =14 642, =12 1187, =10 1999). We stress that this selection simply identifies those stars which are most likely to be variable.
Each source is then subject to a manual inspection of the light curve and corresponding images to verify their variable nature. Where appropriate, additional light curves were obtained using the optimal aperture photometry routine autophotom (Eaton, Draper & Allan 2009). In this process we assign a Flag value (see Table 7) to characterise their lightcurve and timescale of variability. For our =18 sample we found that after a second stage verification process we had 65 sources which we classed as highly likely to be variable using the LS statistic as a first screening stage. A summary of these sources are shown in Table 5 and their colours are shown in Figure 4. By examining samples derived using =14, we have found more than 100 sources which have passed our two stage variable identification process. These are indicated in the FITS file which we make available on the Armagh Observatory Web site.
A high fraction of sources (71 percent) of the =18 sample were found to be unlikely to be bona fide variable sources. Many of these spurious variables were caused by residual systematic trends in the light curves - for instance sources from the same chip showed very similar trends in their light curves. This is almost certainly a result of using manual corrections in the guiding process in the INT observations and for the fact that the light curves only covered a short time (typically 1 hr).
We selected 18 sources which would be good targets to observe using Kepler in 1 min sampling mode. We successfully bid for Kepler Guest Observer Programme and the Directors Discretionary Time Programme. We show the INT light curves of these sources in Figure 5 and we give their sky coordinates, magnitude, colours in Table 2. We indicate an approximate spectral classification using our INT and Gran Telescopio Canarias (GTC) spectra (§4). Of these 18 sources, one is a pulsating DA white dwarf (the details are presented in Greiss et al in prep), three are contact binaries, one is a cataclysmic variable, and one is a flare star (Ramsay et al. 2013). However, most of these sources appear to be Sct stars. We will present an overview of the Kepler data of these sources in §5.
|11911480||19 20 24.9||+50 17 22.4||18.13||–0.39||0.06||12,16||290 sec||DAV (1)|
|3223460||19 12 32.2||+38 23 00.1||13.74||–0.27||0.25||14||24.2 min||GTC||mid-late A||Sct|
|6547396||19 53 18.3||+41 58 26.9||14.84||0.40||0.46||16||26.6 min||INT||mid-late A||Sct|
|8120184||19 54 11.9||+43 59 20.1||14.27||0.36||0.51||15||42.6 min||INT||mid-late A||Sct|
|4377815||19 39 08.1||+39 27 35.9||14.83||0.24||0.34||15||45.7 min||INT||mid-late A||Sct|
|9364179||19 56 24.5||+45 48 24.1||14.38||0.45||0.43||15||46.8 min||INT||mid-late A||Sct|
|9640005||19 09 46.3||+46 20 04.1||18.40||0.15||0.21||14–16||49.5 min||GTC||mid A||Sct|
|8840638||19 55 35.1||+45 04 46.0||14.63||0.52||0.54||14–16||49.6 min||GTC||mid-late A||Sct|
|4636671||19 01 52.2||+39 45 59.3||15.67||0.28||0.26||14–16||50.0 min||GTC||mid A||Sct|
|12406812||19 23 33.8||+51 17 58.9||17.24||0.18||0.36||14–16||50.4 min||GTC||mid-late A||Sct|
|5623923||19 32 01.5||+40 51 16.8||16.62||0.23||0.27||14–16||50.5 min||GTC||mid-late A||EB+ Scuti|
|10284901||19 43 46.4||+47 20 32.8||15.73||0.06||0.32||14,16||75.8 min||GTC||mid-late A||Sct|
|10975348||19 26 46.1||+48 25 30.8||18.89||0.19||0.34||14–16||2.35 hrs||GTC||mid A||Sct|
|7431243||19 08 51.6||+43 00 31.5||19.10||–0.94||0.47||16||4.68 hrs||CV (2)|
|7667885||19 03 30.2||+43 23 22.7||17.64||1.05||0.95||14–16||7.56 hrs||GTC||mid G||W UMa|
|9786165||19 50 11.0||+46 34 40.8||17.67||0.81||0.76||14,16||7.98 hrs||GTC||mid G||W UMa|
|12553806||19 14 41.0||+51 31 08.9||17.52||0.14||0.41||14–16||11.12 hrs||GTC||A+F?||W UMa|
|5474065||19 53 02.5||+40 40 34.6||18.77||1.93||1.43||14||Flare||GTC||M3 V||Flare star (3)|
4 Optical Spectroscopy
As part of our follow up programme, we obtained low–medium resolution optical spectroscopy of over 50 sources which were either identified as being variable on a short timescale or had unusual colours. (One very blue source in our sample, KIC 10449976, has already been reported as an extreme helium star, Jeffery et al. 2013). We obtained data using the Intermediate Dispersion Spectrograph (IDS) and the R400 grism on the INT between 26–28 June 2012 and also using the Optical System for Imaging and Low Resolution Integrated Spectroscopy (OSIRIS) tunable imager and spectrograph and the R1000R grism on the 10.4 m GTC during March – June 2013. Both are located at the Observatorio Roque de los Muchachos in La Palma, Canary Islands, Spain. At least two spectra were obtained of each source and the individual exposure time ranged from 180 to 360 sec (INT) and 40 to 400 sec (GTC). The spectra were reduced using standard procedures with the wavelength calibration being made using a CuNe+CuAr arc taken shortly after the object spectrum was taken. A flux standard was observed so that the (resulting combined) spectra of each source could be flux calibrated in the case of the IDS spectra and to remove the instrumental response in the OSIRIS spectra (the observing programme utilises poor observing conditions). The spectral resolution of our INT spectra was 2Å and 8Å for our GTC spectra.
For stars which were of the spectral type A/F we modelled the spectra using a grid of LTE models calculated using the atlas9 code (Kurucz 1992) with convective overshooting switched off. Spectra were calculated with the linfor line-formation code (Lemke 1991). Data for atomic and molecular transitions were compiled from the Kurucz line list. The stellar temperatures were estimated from the hydrogen Balmer lines of the stars (H to H) using the fitsb2 routine (Napiwotzki et al. 2004). No gravity sensitive features are accessible in our low resolution spectra. McNamara (1997) finds that SX Phe and large amplitude Sct stars have a range in log g of 3.0–4.3. In our fits we fixed log g=4.0, although the resulting temperature is only weakly sensitive to this parameter. The metallicity was allowed to vary although this was not strongly constrained in the fits. The error of the fit parameters were determined with a bootstrapping method. As an example of the fits we show in Figure 6 the fit to the spectrum of KIC 3223460. We show in Table 3 the temperature we derive for the stars which we have obtained Kepler Short Cadence data (§5). We make all the spectra available through the the Armagh Observatory Web site (http://star.arm.ac.uk/rats-kepler) together with the fitted temperature of each.
5 Kepler observations
The detector on board Kepler is a shutterless photometer using 6 sec integrations and a 0.5 sec readout. There are two modes of observation: long cadence (LC), where 270 integrations are summed for an effective 28.4 min exposure, and short cadence (SC), where 9 integrations are summed for an effective 58.8 sec exposure. When an object is observed in SC mode, LC data is also automatically recorded. After the data are corrected for bias, shutterless readout smear and sky background, light curves are extracted using simple aperture photometry (SAP). Data which were contaminated, for instance during intervals of enhanced solar activity, were removed by requiring data to be flagged by the FITS keyword ‘SAP_QUALITY’=0, and the data were corrected for systematic trends (Kinemuchi et al. 2012).
The Kepler data on the pulsating DA white dwarf will be presented in full by Greiss et al. (in prep) while the Kepler data on the flare star KIC 5474065 has been presented by Ramsay et al. (2013). Here we give a brief overview of the Kepler data on the pulsating, contact binaries and cataclysmic variable which we have obtained.
5.1 Sct stars
In Table 2 we identify eleven sources which show a dominant period in the range 24.2 min to 2.35 hrs. Our fits to their low resolution spectra (Table 3) indicate that they have a temperature in the range 7600–8300K. They are therefore consistent with the characteristics of Sct and SX Phe stars (see Breger 2000 for a review). We show a short section (1 day) of each of the Kepler light curves of these Sct type stars in Figure 7. Using a full month of data, we find that their power spectra are complex and show many frequencies (Figure 8).
The Sct stars with the shortest pulsation periods in our sample are KIC 3223460 (24 min) and KIC 6547396 (26 min). Indeed, they are at the extreme short period end of the Sct star distribution (18 mins marks the short period end, Uytterhoeven et al. 2011). The discovery of two Sct stars with such a short period will therefore provide an opportunity to discover the internal structure of these sources through asteroseismology. Eight of our sample, (KIC 8120184, 4377815, 9364179, 9640005, 8840638, 4636671 and 5623923), have a peak in their power spectra which lies in the range 42.6 – 50.5 mins. They have a best fit temperature in the range 7660–7950 K (Table 3). The longest period pulsators, KIC 10284901 (75.8 mins) and 10975348 (2.35 hrs) also show high amplitude variations and appear to be high amplitude scuti stars which occupy a restricted range of the instability strip (McNamara 2000).
Decades of research have shown that the light curves of Sct stars are very complex (e.g. Pamyatnykh 2000, Breger 2000). However, this makes them very useful astrophysical laboratories as they show physical phenomena which can be used to test theoretical models. However, in many Sct stars it is difficult to uniquely identify modes in the power spectra of the light curve. An exception is slow rotators such as 44 Tau (Lenz et al. 2010) where a large variety of pressure and gravity modes were identified. For the majority of these stars, when it comes to modelling their power spectra, a major difficulty is that the mechanism selecting which modes are excited to observable amplitudes is not well understood (Dziembowski & Krolikowska 1990). In other words, some modes are excited while others are not, which makes identifying the specific mode for each peak in the frequency spectra difficult.
The Sct stars are stars with masses between 1.5 and 2.5 and their pulsations are thought to be driven largely by the opacity mechanism in the Heii ionisation zone (Baker & Kippenhahn 1962). From Kepler and CoRoT observations, however, it seems that the opacity mechanism alone cannot excite the entire range of observed modes. This means that either the models are incomplete or that there is an additional mechanism contributing to the driving. Such an alternative explanation would be the presence of stochastically excited modes, like in the Sun. Theoretical models in fact predict the the convective envelopes of Sct stars are still deep and effective enough to drive Solar-like oscillations. Recently, Antoci et al. (2011), suggested the detection of such a hybrid star, showing mechanism and stochastically driven modes. However longer observations revealed that the interpretation is more complicated than initially anticipated (Antoci et al. 2013). It is therefore of great importance to find pulsating stars with similar temperature and gravity to HD 187547 (=7500250 K, log g =3.90.25, Antoci et al. 2011) to test current models. Many of our Sct stars have a similar temperature to HD 187547 but spectra with higher spectral resolution than the ones we present in §4 are required to provide a robust temperature determination. However, even if our Sct stars are likely to be too faint to identify their pulsation modes (even the brightest of our sources at =13.7 is relatively faint for such an analysis), the frequency range and the stability of excited modes can lead to a better understanding of the pulsation mechanisms, provided the temperature is robustly determined.
5.2 Contact Binaries
The Kepler data of three of the sources shown in Table 2 and Figure 5 clearly indicate that they are eclipsing or contact binaries with an orbital period ranging from 0.315 days – 0.463 days. We show the Kepler Q14 SC data of these sources in Figure 9 where we have folded and binned the data on the orbital period.
Prs̆a et al. (2011) and Slawson et al. (2011) present an analysis of the first and second Kepler data release of 4044 eclipsing and contact binaries. Although the shape of the folded light curves of KIC 7667885 and KIC 12553806 are similar to that of semi-detached binaries (also known as Lyr binaries), their relatively short orbital period suggests that they are more likely to be contact binaries (also known as W UMa binaries). The folded light curve of KIC 9786165 also implies it is a contact binary. Although some caution has to be applied in interpreting the results of our spectral fits since we have applied a single temperature model to sources which are clearly binary systems, the temperatures which we derive (Table 3) indicate they are contact binaries rather than semi-detached binaries which have B star components (and hence much hotter).
Unlike the three sources outlined here which have been observed using SC mode, none of the sources shown in Prs̆a et al. (2011) or Slawson et al. (2011) have been observed in SC Mode. Since each of the three binaries have a high inclination, they are excellent data-sets to search for third bodies (such as exo-planets) in these systems.
5.3 Kic 5623923: A Sct star in a contact Binary
We show a 2.5 day section of the light curve of KIC 5623923 in Figure 10. It is clear that this source is an eclipsing or contact binary system with a orbital period of 1.21 days. However, there are clear pulsations on a period of 50 mins superimposed on the light curve. The pulsations are not readily apparent during the secondary eclipses indicating that the secondary star (the less luminous of the binary components) is the source of the pulsations.
There are at least two other eclipsing or contact binary stars in the Kepler field which have a Sct component. KIC 4544587 is a binary system with a 2.18 day orbital period (Hambleton et al. 2013) while KIC 10661783 has an orbital period of 1.23 days (Southworth et al 2011). Unlike KIC 5623923 where the secondary star is the pulsating component, in both KIC 4544587 and KIC 10661783 the primary is the pulsating star. We note that the spectral fits suggest a temperature of 8000 K (Table 3) although we have fitted only a single temperature to a system which is clearly a binary.
The power spectrum of KIC 5623923 (Figure 11) shows many peaks in the 20-30 cycles/day frequency interval which are due to pressure (p) mode pulsations in the secondary star. Some of these peaks are separated by 0.83 cycles/day which is the orbital period. This implies that the amplitude of the Sct pulsations are correlated with the orbital period (see Shibahashi & Kurtz 2012 for a discussion on how power spectra can be used to measure radial velocities in binary systems). The power spectrum of KIC 10661783 shows peaks in its power spectrum in a similar frequency range (Southworth et al. 2011) while the p-modes seen in KIC 4544587 (Hambleton et al. 2013) are at a higher frequency range (40-50 cycles/day). We defer a full analysis of these Kepler data for a dedicated paper.
5.4 KIC 7431243 (V363 Lyr): A Cataclysmic Variable
KIC 7431243 was found to be a moderately blue source in the KIS (=0.47, Greiss et al. 2012) and in our survey it was found to show rapid flux changess superimposed upon an irregular variation (Figure 5). KIC 7431243 matches the variable star V363 Lyr which was discovered as a Cataclysmic Variable (CV) by Hoffmeister (1967), whilst Kato et al (2001) found that it shows outbursts of duration 7–8 days every 21 days. There are several dozen known CVs in the Kepler field (see Scaringi et al. 2013 and Howell et al. 2013).
KIC 7431243 was observed using Kepler in Q16 for 5.2 days and (not surprisingly) no outbursts were seen (Figure 12). The power spectrum of the light curve shows peaks corresponding to 4.68 hrs and 4.47 hrs. If we attribute the longer period to the super-hump period (super-humps are caused by the precession of the accretion disk) and the shorter period to the orbital period, we find the fractional excess, =4.7 percent. Using the relationship of Patterson et al. (2005) this would imply a mass ratio, .
Using the secondary star mass () – orbital period relationship for CVs (, Warner 1995), we find for a CV with =4.47 hrs, =0.42(0.45for =4.68 hrs). Super-humps are thought to be restricted to systems where the mass ratio, , (see Schreiber 2007 for details). If super-humps are present in KIC 7431243 then this may suggest that the white dwarf in this binary has a mass assuming =0.42( for =0.45). Given the potentially high mass of the white dwarf, we urge phase resolved optical spectroscopy this system.
This project set out to identify sources in the Kepler field which showed variability on a timescale of 1 hour or less. The most potentially interesting of these variable sources would then have been subject of bids to obtain Kepler data in Short Cadence. We have identified more than 100 strongly variable sources and we have been succesful in obtaining Kepler SC light curves of 18 of these sources.
Many of them are Scuti stars which show an astonishing range of variability, the star with the shortest dominant period being 24 min. We also identify one Scuti star as being in an eclipsing or contact binary with an orbital period of 1.21 days. As currently only two other such systems are known in the Kepler field, this will provide the means to study binary evolution in more detail. We have also obtained Kepler SC data of three contact binaries and one previously known Cataclysmic Variable. The Kepler observations of one flare star and one pulsating DA white dwarf are reported elsewhere.
We provide a range of images and data products through the Armagh Observatory Web Site (star.arm.ac.uk/rats-kepler). These include the reduced images so that users can perform photometric measurements using their favoured reduction packages. We also provide the detrended light curves and the photometric variability parameters of each source observed in our survey.
The Isaac Newton Telescope is operated on the island of La Palma in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofisica de Canarias (IAC) with financial support from the UK Science and Technology Facilities Council. We would like to thank the ING and MDM staff for their support. Observations were also made with the Gran Telescopio Canarias (GTC) which is also sited on La Palma and run by the IAC. Armagh Observatory is supported by the Northern Ireland Government through the Department of Culture, Arts and Lesuire. We thank Wojtek Pych for the use of his difference imaging software diapl2. DS acknowledges support of STFC through an Advanced Fellowship. Funding for the Stellar Astrophysics Centre (Aarhus) is provided by The Danish National Research Foundation. The research is supported by the ASTERISK project (ASTERoseismic Investigations with SONG and Kepler) funded by the European Research Council (Grant agreement no.: 267864). We thank the referee for helpful comments which helped to significantly improve the paper.
Appendix A Tables
|Date||Field ID||RA DEC||Date(dd-mm-yy)||Field ID||RA DEC (J2000)|
|11-07-11||522||19:46:51 +49:43:35||02-08-11||382||19:22:42 +50:39:03|
|11-07-11||542||19:45:05 +47:36:21||02-08-11||408||19:23:33 +49:11:06|
|11-07-11||541||19:46:43 +47:14:21||03-08-11||46||19:01:23 +40:59:18|
|11-07-11||349||19:39:18 +39:16:51||03-08-11||63||19:05:58 +37:58:36|
|11-07-11||539||19:49:56 +46:30:21||03-08-11||260||19:11:38 +46:19:02|
|11-07-11||567||19:57:19 +43:59:07||03-08-11||73||19:11:46 +38:20:36|
|11-07-11||570||19:52:40 +45:05:09||03-08-11||259||19:13:14 +45:57:02|
|12-07-11||94||18:44:31 +47:36:58||03-08-11||380||19:14:02 +51:34:03|
|12-07-11||100||18:45:45 +48:09:58||03-08-11||373||19:14:10 +50:39:03|
|12-07-11||93||18:46:25 +47:14:58||04-08-11||72||19:02:36 +39:37:36|
|12-07-11||234||18:57:27 +49:21:54||04-08-11||236||19:06:53 +48:26:54|
|12-07-11||575||19:55:42 +45:16:09||04-08-11||264||19:16:22 +46:08:02|
|12-07-11||503||19:53:20 +42:09:20||04-08-11||296||19:18:56 +45:13:52|
|12-07-11||488||19:53:27 +40:19:20||04-08-11||277||19:19:29 +47:14:02|
|13-07-11||99||18:47:40 +47:47:58||04-08-11||269||19:19:33 +46:19:02|
|13-07-11||105||18:48:54 +48:20:58||04-08-11||406||19:26:55 +48:27:05|
|13-07-11||92||18:48:18 +46:52:58||05-08-11||245||19:02:50 +50:05:54|
|13-07-11||157||19:15:34 +42:19:19||05-08-11||76||19:07:31 +39:26:36|
|13-07-11||469||19:43:31 +44:27:24||05-08-11||281||19:20:36 +43:01:58|
|13-07-11||446||19:40:23 +46:44:30||05-08-11||223||19:28:21 +39:15:28|
|13-07-11||354||19:42:06 +39:27:51||05-08-11||325||19:36:07 +41:32:30|
|14-07-11||104||18:50:50 +47:58:58||05-08-11||583||20:06:28 +44:32:07|
|14-07-11||97||18:51:29 +47:03:58||05-08-11||585||20:03:23 +45:16:09|
|14-07-11||103||18:52:45 +47:36:58||06-08-11||578||20:01:52 +44:43:09|
|14-07-11||110||18:54:01 +48:09:58||06-08-11||555||19:59:31 +47:03:21|
|14-07-11||204||19:24:20 +37:36:28||06-08-11||499||19:59:13 +40:41:20|
|14-07-11||214||19:29:50 +37:58:28||06-08-11||566||19:58:51 +43:37:09|
|14-07-11||213||19:31:14 +37:36:28||06-08-11||580||19:58:44 +45:27:07|
|15-07-11||8||18:46:20 +42:51:19||06-08-11||588||19:58:37 +46:22:09|
|15-07-11||7||18:48:05 +42:29:19||06-08-11||556||19:57:53 +47:25:21|
|15-07-11||12||18:51:04 +42:40:19||07-08-11||581||19:57:09 +45:49:09|
|15-07-11||18||18:52:17 +43:13:19||07-08-11||494||19:54:53 +40:52:20|
|15-07-11||321||19:31:44 +41:43:30||07-08-11||550||19:54:41 +47:14:21|
|15-07-11||297||19:28:09 +43:56:55||07-08-11||562||19:54:17 +43:48:09|
|15-07-11||328||19:31:40 +42:38:30||07-08-11||559||19:52:56 +48:31:21|
|16-07-11||114||18:57:18 +44:04:28||07-08-11||563||19:52:45 +44:10:09|
|16-07-11||120||18:58:32 +44:37:28||07-08-11||471||19:52:37 +43:10:24|
|16-07-11||126||18:59:46 +45:10:28||08-08-11||538||19:51:32 +46:08:21|
|16-07-11||314||19:31:45 +40:48:30||08-08-11||560||19:51:16 +48:53:21|
|16-07-11||309||19:28:51 +40:37:30||08-08-11||483||19:50:35 +40:08:20|
|16-07-11||338||19:35:06 +38:32:48||08-08-11||490||19:50:33 +41:03:20|
|16-07-11||219||19:34:00 +37:47:28||08-08-11||497||19:50:28 +41:58:18|
|17-07-11||28||18:48:05 +44:52:19||08-08-11||528||19:50:12 +49:54:35|
|17-07-11||27||18:49:54 +44:30:19||08-08-11||465||19:49:37 +42:59:24|
|17-07-11||133||18:59:24 +46:05:28||09-08-11||473||19:49:34 +43:54:24|
|17-07-11||55||18:59:05 +42:16:18||09-08-11||477||19:49:10 +39:35:20|
|17-07-11||412||19:28:28 +49:00:06||09-08-11||484||19:49:09 +40:30:20|
|17-07-11||436||19:32:35 +46:44:27||09-08-11||521||19:48:33 +49:21:35|
|17-07-11||439||19:38:53 +46:11:30||09-08-11||533||19:48:22 +45:57:19|
|01-08-11||36||18:55:39 +40:37:18||09-08-11||540||19:48:20 +46:52:21|
|01-08-11||238||19:02:58 +49:10:54||09-08-11||466||19:48:07 +43:21:23|
|01-08-11||139||19:02:58 +46:16:28||10-08-11||530||19:46:45 +50:38:35|
|01-08-11||131||19:03:05 +45:21:28||10-08-11||548||19:46:36 +48:09:21|
|01-08-11||376||19:21:01 +50:06:03||10-08-11||467||19:46:35 +43:43:23|
|02-08-11||240||18:59:00 +49:54:54||10-08-11||486||19:46:14 +41:14:17|
|02-08-11||151||19:03:09 +43:25:19||10-08-11||516||19:45:13 +49:10:35|
|02-08-11||141||19:08:12 +41:24:19||10-08-11||535||19:45:11 +46:41:21|
|02-08-11||154||19:09:37 +42:52:19||10-08-11||454||19:45:09 +42:15:24|
|Date||Field ID||RA DEC||Date||Field ID||RA DEC (J2000)|
|03-08-12||5||18:39:52 +43:24:19||08-08-12||1010||19:20:19 +43:38:30|
|03-08-12||67||18:59:22 +39:26:36||08-08-12||203||19:25:43 +37:14:28|
|03-08-12||144||19:03:15 +42:30:19||08-08-12||251||19:06:41 +50:16:54|
|03-08-12||300||19:23:31 +45:02:56||08-08-12||289||19:19:02 +44:18:58|
|05-08-12||463||19:42:05 +43:54:24||08-08-12||383||19:20:57 +51:01:03|
|05-08-12||1001||18:45:00 +47:21:36||09-08-12||1005||19:12:34 +43:30:14|
|05-08-12||1007||18:59:02 +48:42:37||09-08-12||1006||19:17:19 +39:27:18|
|05-08-12||155||19:08:06 +43:14:17||10-08-12||1006||19:17:19 +39:27:18|
|05-08-12||568||19:55:46 +44:21:07||10-08-12||1008||19:11:33 +45:43:44|
|05-08-12||463||19:42:05 +43:54:24||10-08-12||210||19:25:41 +38:09:28|
|05-08-12||546||19:49:53 +47:25:22||10-08-12||222||19:29:47 +38:53:28|
|05-08-12||523||19:45:08 +50:05:35||10-08-12||365||19:15:57 +49:22:03|
|06-08-12||1||18:46:52 +41:56:18||10-08-12||368||19:10:48 +50:28:03|
|06-08-12||1002||19:04.62 +42:45:48||10-08-12||374||19:12:25 +51:01:03|
|06-08-12||168||19:10:56 +44:20:19||11-08-12||1009||19:09:59 +47:17:07|
|06-08-12||175||19:17:49 +39:37:14||11-08-12||1011||19:18:30 +45:33:11|
|06-08-12||181||19:19:14 +40:10:15||11-08-12||1012||19:19:12 +49:57:51|
|06-08-12||305||19:28:07 +44:52:02||11-08-12||292||19:25:07 +43:45:58|
|06-08-12||342||19:39:19 +38:21:51||11-08-12||359||19:46:21 +39:16:51|
|07-08-12||1013||19:29:12 +50:19:02||11-08-12||392||19:19:00 +52:18:03|
|07-08-12||226||19:00:03 +48:04:54||11-08-12||401||19:23:41 +48:16:06|
|07-08-12||239||19:00:59 +49:32:54||12-08-12||10||18:42:48 +43:35:20|
|07-08-12||311||19:25:55 +41:21:30||12-08-12||2||18:45:08 +42:18:20|
|07-08-12||327||19:33:09 +42:16:30||12-08-12||3||18:43:23 +42:40:20|
|07-08-12||332||19:37:34 +42:05:30||12-08-12||4||18:41:38 +43:02:20|
|07-08-12||351||19:36:26 +40:00:51||12-08-12||496||19:51:57 +41:36:20|
|08-08-12||1003||19:06:31 +43:54:48||12-08-12||6||18:49:50 +42:07:20|
|08-08-12||1004||19:19:55 +42:47:29||12-08-12||9||18:44:35 +43:13:20|
|Date||Field ID||RA DEC||Date||Field ID||RA DEC|
|16-05-2012||81||19:12:07 +39:16:50||21-05-2012||174||19:19:15 +39:15:15|
|16-05-2012||82||19:10:42 +39:38:50||21-05-2012||175||19:17:49 +39:37:15|
|16-05-2012||83||19:09:16 +40:00:51||21-05-2012||226||19:00:03 +48:04:55|
|16-05-2012||84||19:07:49 +40:22:55||21-05-2012||351||19:36:26 +40:00:51|
|17-05-2012||85||18:49:13 +45:58:48||22-05-2012||212||19:22:51 +38:53:28|
|19-05-2012||135||19:10:10 +44:50:04||22-05-2012||218||19:24:11 +39:26:28|
|19-05-2012||142||19:07:08 +41:47:26||22-05-2012||273||19:13:03 +47:47:02|
|19-05-2012||149||19:07:04 +42:42:34||22-05-2012||95||18:42:35 +47:58:58|
|19-05-2012||89||18:41:41 +47:26:48||23-05-2012||106||18:46:58 +48:42:58|
|20-05-2012||122||18:54:54 +45:21:28||23-05-2012||286||19:23:37 +43:12:58|
|20-05-2012||160||19:11:03 +43:25:19||23-05-2012||288||19:20:34 +43:56:58|
|20-05-2012||161||19:09:32 +43:47:19||23-05-2012||329||19:30:09 +43:00:30|
|KIC_ID||The Kepler Input Catalog (Brown et al. 2011) star number;|
|RA, DEC||Right Ascension and Declination (J2000);|
|Taken from the KIS, Greiss et al. (2012a,b);|
|Taken from the KIS Greiss et al. (2012a,b);.|
|U_g, g_r||, taken from Greiss et al. (2012a,b);|
|LS_Period||The period of the most prominent peak in the Lomb Scargle periodogram in days and LS_Period_Mins (mins);|
|Log10_LS_Prob||The False Alarm Probability of the most prominent period in the Lomb Scargle periodogram (in units of log 10);|
|Alarm||The alarm variability statistic (Tamuz, Mazeh, and North 2006);|
|AoV_Period||The Period determined from the AoV test in days and AOV_Period_Mins (mins);|
|AoV||The AoV variability statistic Schwarzenberg-Czerny (1989, 1996);|
|AOV_SNR||the S/N ratio of the peak measured over the full periodogram;|
|AOV_NEG_LN_FAP||The negative of the natural logarithm of the formal false alarm probability;|
|Chi2||The reduced value of the light curve tested against the constant mean value (with 5 clipping);|
|StdDev||the standard deviation (root mean squared) of the light curve;|
|Field||Our internal naming convention for the field pointing;|
|Chip||For the INT/WFC the chip id. There was only one chip for the MDM observations;|
|FieldChip||The Field-Chip Combination|
|ID||Our internal naming convention for the source. A 6 digit number implies the light curve was derived;|
|using diapl, while for those derived using sextractor, the numbering system starts from 1;|
|X,Y||The X,Y coordinates of the source on the chip;|
|grats, g_r_rats||The mag and colour at the time of our observations;|
|Flag||‘0’ Variability on ls_per_min; ‘1’ Probable variability on ls_per_min; ‘2’ Clear long timescale high amplitude variable;|
|‘3’ Not variable on ls_per_min; ‘4’ Variable on period other than ls_per_min; ‘5’ Possibile variability in genera;l|
|‘6’ not likely to be variable; ‘7’ bad light curve; ‘8’ Eclipse; ‘9’ Possible eclipse;|
|‘10’ Variability likely due to systematic trend; ‘11’ Known bad columns on chip;|
|‘12’ Apparent long period could be due to residual systematic trends; ‘13’ Image shows close stellar companion;|
|Tstart, Tstop||The start and end date in MJD of the sequence of band observations.|
|medFAP||The median log FAP for the chip which the source is located|
- Alard, C., Lupton, R. H., 1998, ApJ, 503, 325
- Antoci V., et al., 2011, Nature, 477, 570
- Antoci, V., et al., 2013, in press, MNRAS
- Baker N., Kippenhahn R., 1962, ZA, 54, 114
- Baran A. S., Gilker J. T., Fox-Machado L., Reed M. D., Kawaler S. D., 2011, MNRAS, 411, 776
- Barclay T., Ramsay G., Hakala P., Napiwotzki R., Nelemans G., Potter S., Todd I., 2011, MNRAS, 413, 2696
- Bedding, T. R., et al., 2011, Nature, 471, 608
- Bertin E., Arnouts S., 1996, A&AS, 117, 393
- Breger, M., 2000, ASP Con Series, 210, 3
- Brown T. M., Latham D. W., Everett M. E., Esquerdo G. A., 2011, AJ, 142, 112
- Chaplin W. J., et al., 2011, Sci, 332, 213
- Devor, J., 2005, ApJ, 628, 411
- Dziembowski, W.; Krolikowska, M., 1990, AcA, 40, 19
- Eaton, N., Draper, P. W., & Allan, A., 2009, Starlink User Note, 45
- Feldmeier J. J., et al., 2011, AJ, 142, 2
- Graham, M. J., Drake, A. J., Djorgovski, S. G., Mahabal, A. A., Donalek, C., Duan, V., Maher, A., 2013, MNRAS, 434, 3423
- Greiss S., et al., 2012a, AJ, 144, 24
- Greiss S., et al., 2012b, arXiv:1212.3613
- Groot, P. J., et al., 2009, MNRAS, 399, 323
- Hambleton, K. M., et al, 2013, MNRAS, 434, 925
- Hartman J. D., Bakos G., Stanek K. Z., Noyes R. W., 2004, AJ, 128, 1761
- Hartman J. D., Gaudi B. S., Holman M. J., McLeod B. A., Stanek K. Z., Barranco J. A., Pinsonneault M. H., Kalirai J. S., 2008, ApJ, 675, 1254
- Hoffmeister, C., 1967, AN, 289, 205
- Howell, S. B., Everett, M. E., Seebode, S. A., Szkody, P., Still, M., Wood, M., Ramsay, G., Cannizzo, J., Smale, A., 2013, AJ, 145, 109
- Jeffery C. S., et al., 2013, MNRAS, 429, 3207
- Jester, S., et al, 2005, AJ, 130, 873
- Kato, T., Nogami, D., Baba, H., Masuda, S., 2001, IBVS, 5118
- Kinemuchi, K., Barclay, T., Fanelli, M., Pepper, J., Still, M., Howell, S. B., 2012, PASP, 124, 963
- Koch D. G., et al., 2010, ApJ, 713, L79
- Kurucz, R. L., 1992, Rev Mex Astron Astro., 23, 181
- Lang D., Hogg D. W., Mierle K., Blanton M., Roweis S., 2010, AJ, 139, 1782
- Lemke, M. 1991, Internal Report, Department of Astronomy, University of Texas at Austin
- Lenz P., Pamyatnykh A. A., Zdravkov T., Breger M., 2010, A&A, 509, A90
- Lomb, N.R. 1976, A&SS, 39, 447
- McNamara, D. H., 1997, PASP, 109, 1221
- McNamara, D. H., 2000, Delta Scuti and Related Stars, Reference Handbook and Proc 6th Vienna Workshop in Astrophysics, ASPC, 210, 373
- Napiwotzki, R., et al, 2004, In Spectroscopically and Spatially Resolving the Components of the Close Binary Stars, ASP Conf Series, 318, 402
- Pamyatnykh, A. A., Delta Scuti and Related Stars, Reference Handbook and Proc 6th Vienna Workshop in Astrophysics, ASPC, 210, 215
- Patterson, J., et al. 2005, PASP, 117, 1204
- Pigulski A., Pojmański G., Pilecki B., Szczygieł D. M., 2009, AcA, 59, 33
- Pinsonneault M. H., An D., Molenda-Żakowicz J., Chaplin W. J., Metcalfe T. S., Bruntt H., 2012, ApJS, 199, 30
- Prs̆a, A., et al., 2011, AJ, 141, 83
- Ramsay G., Hakala P., 2005, MNRAS, 360, 314
- Ramsay, G., Napiwotzki, R., Hakala, P., Lehto, H., 2006, MNRAS, 371, 957
- Ramsay, G., Napiwotzki, R., Barclay, T., Hakala, P., Potter, S., Cropper, M., 2011, MNRAS, 417, 400
- Ramsay, G., Doyle, J. G., Hakala, P., Garcia-Alvarez, D., Brooks, A., Barclay, T., Still, M., 2013, MNRAS, 434, 2451
- Scargle, J.D. 1982, ApJ, 263, 835
- Scaringi, S., Groot, P. J., Verbeek, K., Greiss, S., Knigge, C., Körding, E., 2013, MNRAS, 428, 2207
- Schwarzenberg-Czerny A., 1989, MNRAS, 241, 153
- Schwarzenberg-Czerny A., 1996, ApJ, 460, L107
- Shibahashi, H., Kurtz, D. W., 2012, MNRAS, 422, 738
- Slawson, R. W., et al, 2011, AJ, 142, 160
- Southworth, J., et al, 2011, MNRAS, 414, 2413
- Tamuz O., Mazeh T., Zucker S., 2005, MNRAS, 356, 1466
- Tamuz O., Mazeh T., North P., 2006, MNRAS, 367, 1521
- Thompson, S. E., et al., 2012, ApJ, 753, 86
- Uytterhoeven K., et al., 2011, A&A, 534, A125
- Warner, B. 1995, Cataclysmic Variable Stars (Cambridge: Cambridge)
- Welsh W. F., et al., 2011, ApJS, 197, 4
- Wozniak P. R., 2000, AcA, 50, 421