ap-6-11.dvi Acta Polytechnica Vol. 51 No. 6/2011 Photometric Analysis of the Pi of the Sky Data M. Siudek, K. Malek, L. Mankiewicz, R. Opiela, M. Sokolowski, A. F. Żarnecki Abstract A database containing star measurements from the period 2006–2009 taken by the Pi of the Sky detector located in Las Campanas Observatory in Chile contains more than 2 billion measurements of almost 17 million objects. All measurements are available on the Pi of the Sky web site through a dedicated interface, which also allows users to download selected data. Accurate analysis of Pi of the Sky data is a real challenge, because of a number of factors that can influence the measurements. Possible sources of errors in our measurements include: reading the chip with the shutter open, strong and varying sky background, passing planets or planetoids, and clouds and hot pixels. In order to facilitate the analysis of variable stars we have developed a system of dedicated filters to remove bad measurements or frames. The spectral sensitivity of the detector is taken into account by appropriate corrections based on the spectral type of reference stars. This process is illustrated by an analysis of the BG Ind system, where we have been able to reduce the systematic uncertainty to about 0.05 magnitudo. Keywords: Gamma Ray Burst (GRB), prompt optical emissions, optical flashes, nova stars, variable stars, robotic telescopes, photometry. 1 Introduction Pi of the Sky [4] is a robotic telescope aimed at monitoring a large part of the sky with a good range and time resolution. The detector was designed for observations of astrophysical phenomena character- izedby short timescales, especially forpromptoptical counterparts of Gamma RayBursts (GRBs). The Pi of the Sky apparatus design and observational strat- egy make this detector useful also for searching for nova and supernova stars and for monitoring inter- esting objects such as blasars, AGNs and variable stars. The full Pi of the Sky systemwill consist of 2 sites separated by a distance of about 100 km, allowing flashes from satellites and other near-Earth objects to be rejected. Each site will consist of 12 custom- designed CCD cameras placed on specially designed equatorial mounts (4 cameras per mount). The full system is now under construction. Necessary tests before constructing the final ver- sion were performed with a prototype consisting of 2 custom-designed cameras placed on an equatorial mount. The prototype operated at the Las Cam- panas Observatory in Chile from June 2004 till the end of 2009. Cameras working in coincidence ob- served a 20◦ × 20◦ field of view with a time resolu- tion of 10 seconds. Each camera is equipped with Canon lenses f = 85 mm, d = f /1.2, which allows observation of objects to about 12m (about 13.5m for 20 coadded frames). During the 6 years of work the prototype gathered a large amount of data valuable for astronomical research, e.g. for identifying and cataloging many different types of variable stars [2]. 2 Pi of the Sky databases All data gathered by the Pi of the Sky detector are stored in publicly accessible databases1. Currently three databases are available: the first database, covering the period from June 2004 till May 2005, contains almost 800 million measurements for about 4.5 million objects. The second database, from May 2006–November 2007, is a subset of the third database, which contains data gathered from May 2006–April 2009. The third database is the biggest, and includes about 2.16 billion measurements for about 16.7 million objects. In order to ensure easy access to our measurements, we have developed a user-friendly web interface [1]. It allows stars to be selected according to their type, magnitudo, coor- dinates, etc, displaying their light curves and other properties. It is also possible to download largepack- ets of light curves of multiple stars. 3 Improving photometry 3.1 System of dedicated filters The data acquired during the Pi of the Sky ob- servations is reduced by a fully automatic pipeline, and only the light curves of stars are stored in the database. In many aspects of analysis, e.g. identifi- cation of variable stars, it is important to select only 1Pi of the Sky Home Page: http://grb.fuw.edu.pl/pi/ 68 Acta Polytechnica Vol. 51 No. 6/2011 Fig. 1: Precision dispersion of star brightness measurements from standard photometry for 200s exposures (20 coad- ded frames) from the Pi of the Sky prototype at the Las Campanas Observatory in Chile. Large dispersion (left) is mainly caused by false measurements. After the application of quality cuts (right), the photometry accuracy improves significantly data with high measurement precision. In order to fulfill this requirement we have developed a system of data quality cuts. The system of dedicated filters allows the rejection of measurements or even whole frames affected by detector imperfections or observa- tion conditions. Measurements that are placed near the border of the frame, or that are infected by hot pixels, bright background caused by an open shutter or moon halo, or by planet or planetoid passage, can easily be excluded from further analysis. Filters al- lowing the removal of measurements infected by one of these effects are available on the web interface. Applying quality cuts can significantly improve photometry accuracy (see Figure 1). For stars with range 7–10m, average photometry uncertainty sigma of about 0.018–0.024m has been achieved. 3.2 Approximate color calibration algorithm The Pi of the Sky detector is characterized by rela- tively wide spectral sensitivity due to the fact that the observations are performed only with an IR-UV- cut filter. The average wavelength is about 585 nm and corresponds to a V filter, which is also used as a reference in photometry corrections. It turned out that the detector response is correlatedwith the star spectral type. The average magnitudo measured by Pi of the Sky (MP I) is shiftedwith respect to the dif- ference of the catalogue magnitudo given by B-V or J-K. Approximativy of this dependence with a lin- ear function enables the measurement of each star to be corrected. After applying the correction, the distribution of the average magnitudo shift for the reference stars becomes significantlynarrower (an ex- ample of the reference stars for BG Ind variable [3] is shown in Figure 2). Fig. 2: Distribution of the average magnitudo shift for reference stars after standard photometry (red) and after spectral correction (blue) Further conditions, including maximal accepted shift, RMS of Mcoor and number of measurements, are used to evaluate additional photometry correc- tions (see Figure 3). Fig. 3: For calculating photometry corrections, only the best reference stars were used (blue) after rejecting stars with magnitudo shift (Mcoor − V ) bigger than 0.2, RM Scoor bigger than 0.07 andwith number of measurements smaller than 100 (red) 69 Acta Polytechnica Vol. 51 No. 6/2011 Significant improvement of measurement preci- sion is also achievedwhen the photometry correction is not calculated as a simple average over all selected reference stars, but when a quadratic dependence of the correction on the reference star position in the sky is fitted for each frame. χ2 distribution can be used to select measurements with the most precise photometry (see Figure 4). The effect of photometry correctionwith a distri- bution of χ2 on the reconstructedBG Ind light curve is shown in Figure 5. Applying the new algorithm improved the pho- tometry quality, and uncertainty sigma of the order of 0.013m was obtained [3]. We also applied photom- etry correction to other stars, with as good results as in the case of BG Ind variable. Spectral correc- tionandadditional χ2 distributionallowthe selection of only the measurements with the highest precision. With the example of cepheid variable,TCruandGH Carwecan see that the light curve is significantly im- proved and the measurements flagged as those with potentially worse precision do indeed tend to stand outside the light curve (see Figure 6). Fig. 4: χ2 distribution. For about 20 % of the calcu- lated frames, χ2 is greater than 0.058. This information can beused to selectmeasurementswith themost precise photometry Fig. 5: Uncorrected light curve for BG Indvariable (left) and after spectral correctionswith correction quality cut (right) Fig. 6: Uncorrected light curve for T Cru variable (left) and after spectral corrections with correction quality cut (right) Fig. 7: Uncorrected light curve for GH Car variable (left) and after spectral corrections with correction quality cut (right) 70 Acta Polytechnica Vol. 51 No. 6/2011 4 Summary During the period 2006–2009 the prototype observed almost 17 million objects and collected over 2 bil- lion measurements. The data acquired by Pi of the Sky is publicly available through a user-friendly web interface. In order to facilitate selection of measure- ments with high precision, we have developed a sys- tem of dedicated filters to remove measurements af- fected by detector imperfections or observation con- ditions. When an approximate color calibration al- gorithmbased on the spectral type of reference stars is used, an uncertainty sigma of the order of 0.013m can be obtained. Acknowledgement We are very grateful to G. Pojmanski for access to the ASAS dome and for sharing his experience with us. We would like to thank the staff of the LasCam- panas Observatory, San Pedro de Atacama Obser- vatory and the INTA El Arenosillo test centre in Mazagón near Huelva for their help during instal- lation and maintenance of our detector. This work was financed by the Polish Ministry of Science and HigherEducation in 2009–2011as a researchproject. References [1] Biskup, M., et al.: Web interface for star databases of the Pi of the Sky Experiment, Pro- ceedings of SPIE, Vol. 6937, 2007. [2] Majczyna, A., et al.: Pi of the sky catalogue of the variable stars from 2006–2007 data, Proceed- ings of SPIE, Vol. 7745, 2010. [3] Rozyczka,M.,Malek,K., et al.: Absolute proper- ties of BG Ind — a bright F3 system just leaving theMain Sequence,Monthly Notices of the Royal Astronomical Society, 2011, 414, 2479–2485. [4] Siudek, M., et al.: Pi of the Sky telescopes in Spain and Chile, in these proceedings. Malgorzata Siudek Katarzyna Malek Lech Mankiewicz Rafal Opiela Center for Theoretical Physics of the Polish Academy of Sciences Al. Lotnikow 32/46, 02-668 Warsaw, Poland Marcin Sokolowski The Andrzej Soltan Institute for Nuclear Studies Hoza 69, 00-681 Warsaw, Poland Aleksander F. Żarnecki Faculty of Physics University of Warsaw Hoza 69, 00-681 Warsaw, Poland 71