1. Pi of the Sky – robotic search for GRBs
T. Batscha, H. Czyrkowskib, A. Ćwieka, M. Ćwiokb, R. Dąbrowskib, G. Kasprowiczc, A. Majchera,
A. Majczynaa, K. Małekd,e, L. Mankiewicze, K. Nawrockia, Ł. Obarab, R. Opielae, L. W. Piotrowskib, M. Siudeke,
M. Sokołowskif,g,a, R. Wawrzaszekh, G. Wrochnaa, M. Zarembab, A. F. Żarneckib
a) National Centre for Nuclear Research, ul. Andrzeja Sołtana 7, Otwock, Świerk
b) Faculty of Physics University of Warsaw, ul. Hoża 69,
c) Institute of Electronic Systems Faculty of Electronics and Information Technologies Warsaw University of Technology, ul Nowowiejska 15/19, Warszawa
d) Department of Particle and Astrophysical Science, Nagoya University, Furo-cho, Chikusa-ku, Nagoya , Japan
e) Center for Theoretical Physics PAS, Al. Lotników 32/46, Warszawa
f) International Centre for Radio Astronomy Research - Curtin University, GPO Box U1987, Perth, WA 6845, Australia
g) ARC Centre of Excellence for All-sky Astrophysics (CAASTRO)
h) Space Research Center PAS, ul. Bartycka 18A, Warszawa
Pi of the Sky is a system of robotic telescopes designed for observations of short timescale astrophysical phenomena, like prompt optical GRB emission. The apparatus is designed to monitor a large fraction of the sky with 12m-13m range and time resolution of the order of s.
In October 2010 the first unit of the new Pi of the Sky detector system was successfully installed in the INTA El Arenosillo Test Centre in Spain. We have also successfully moved our prototype detector from Las Campanas Observatory to San Pedro de Atacama Observatory in March Soon, 3 more units (12 CCD cameras) covering together 4800 square degrees, will be installed on a new platform in INTA.
Photometric corrections with BG Ind as exemplification
Lightcurves obtained from standard data reduction (background subtraction and flat division) usually are affected by different sort of distortions, due to reading the chip with the opened shutter, strong and varying sky background, passing planets or planetoids, clouds or hot pixels (see Fig. 3).
INTA (near Huelva, Spain)
SPdA (San Pedro de Atacama, Chile)
Fig. 3. Phased BG Ind lightcurve without any correction. Period P = day.
Fig. 4. Phased BG Ind lightcurve with spectral correction applied.
Another important source of uncertainty is, that we perform observations without any filter (except one camera in Chile), but we normalize our measurements into V filter catalogues. And thus we are sensitive to the spectral type of reference stars. For this reason an additional correction was introduced for reference stars, in which we use brightnesses in J and K filters:
Mcorr = M •(J - K),
where: M – apparent magnitude, J, K – brightnesses in J and K filters respectively, taken from 2MASS catalogue.
With the proper sellection of reference stars (in 5o radius from the investigated star and with brightness between 6-10 mag) we were able to obtain a photometric error σ ∽ mag. In Fig. 4 the lightcurve of BG Ind after the application of improved photometry procedures is shown (Kałużny et al. 2011).
- four cameras on equatorial mount,
- field of view 40ox40o (WIDE mode) or 20ox20o (DEEP mode),
- UV – IR cut filters,
- two cameras on equatorial mount,
- field of view 20ox20o,
- one camera has an R (Johnson Bessel) filter installed,
- Canon photolenses,
- focus f = 85 mm, lens speed f/d = 1.2,
- matrix size = 2048x2048 pixels,
- pixel size 15x15 μm2 (36”).
Selected results
GRB B
March 19th 2008, at 6:12:49 UT automatic algorithms detected a new object in the sky. A few seconds later an alert from GCN arrived - the Swift satellite detected an extremely luminous GRB. Also in optical band its brightness was greater than all bursts observed until then. In maximum it was as bright as 5.3 mag. After four and a half year this record still holds.
Measurements of a parallax
Every night, when both places observe, automatic algorithms search for new objects in the sky, especially GRB, in optical band. Unfortunately most flashes are due to satellites or planes, but there is also a number of cases which are difficult or impossible for unique identification. The best method to recognize astrophysical sources is to measure distance to them. And the best method to do it is parallax, which is possible to measure by us, since we have two observatories. The base is
The most curious was, that luminosity measured by Pi of the Sky was over times greater, than luminosity extrapolated from gamma to optical band. This shows, that optical emission is caused by different mechanism, than emission in gamma rays (Racusin et al. 2008).
Fig. 1. Optical lightcurve of GRB B.
Fig. 5. Differences between distances to satellites calculated and taken from TLE base. There is a correction to Earth's radius and 1000 km cut.
Search for the GRB080319B optical precursor
To meet the requirement for monitoring a large fraction of the sky, the Pi of the Sky apparatus makes use of cameras with a wide field of view. For stars far from the optical axis, this causes significant deformations of images, which are much larger than in other astronomical experi-ments. This was also the case for GRB080319B, for which the position of the burst was in the corner of the frame up to t s. The possible precursor would therefore also be deformed and thus large uncertainties would be introduced into standard
huge, almost 8500 km, and we are able to meausre parallax angles between 25" and 14o. Combination of those conditions results, we are able to measure distances between about km and 38 million km from the Earth centre. In this range one can measure geostationary and GPS satellites, space debris, also (very rarely) comets and planetoids. In the future we hope to measure a parallax to closer objects, if second observatory in Spain (near Malaga, 240 km from INTA) will be open.
In Fig. 5 there are differences between distances to satellites calculated from observations of parallax (corrected to observatories localization) and taken from TLE database. Differences smaller than 1000 km are shown. From single observation we are able to measure distances to satellites with 50 km accuracy.
Fig. 2. An example of PSF near the frame corner.
         photometric and signal-searching algorithms.
To improve measurements and signal seeking capabilities a model of the Point Spread Function (PSF), based on modified Zernike polynomials, in the Pi of the Sky system was created (L.W.Piotrowski et al. 2013). Simulated PSFs obtained from the model are very close to real images even for the most deformed stars, as can be seen in fig. 2. The precursor search was performed by fitting the PSF model at GRB coordinates to all frames covering 19 minutes prior to the explosion, on two cameras of the Pi of the Sky prototype. No signal exceeding 3σ level has been found. The limit calculated on single frame fluctuated in most cases between 11.5m and 12.25m (L. W. Piotrowski, 2012).
Acknowledgments
This work is partially financed by:
- grant of Polish Ministry of Science and Higher Education in
- grant number N N (AM).
- POLISH-SWISS ASTRO PROJECT cofound under the Swiss programm of
cooperation with new member states of European Union (AM, AĆ).