------------------------------------------------------------------------ From paolo@discovery.saclay.cea.fr Thu Sep 11 04:23:41 1997 Date: Thu, 11 Sep 1997 10:23:20 +0200 (MET DST) From: Paolo Goldoni X-Sender: paolo@soyouz To: gcnews@astro.umd.edu Subject: MIME-Version: 1.0 %astro-ph/9709096 \documentstyle[epsfig,longtable]{aipproc} \begin{document} \title{The IBIS view of the galactic centre: INTEGRAL's imager observations simulations} \author{P. Goldoni, A. Goldwurm, P. Laurent, F. Lebrun} \address{CEA/DSM/DAPNIA, SAp CEA-Saclay F-91191 Gif-sur-Yvette FRANCE} \maketitle \begin{abstract} The Imager on Board Integral Satellite (IBIS) is the imaging instrument of the INTEGRAL satellite, the hard-X/soft-gamma ray ESA mission to be launched in 2001. It provides diagnostic capabilities of fine imaging (12' FWHM), source identification and spectral sensitivity to both continuum and broad lines over a broad (15~keV--10~MeV) energy range. It has a continuum sensitivity of 2~10$^{-7}$~ph~cm$^{-2}$~s$^{-1}$ at 1~MeV for a 10$^6$ seconds observation and a spectral resolution better than 7~$\%$ at 100~keV and of 6~$\%$ at 1~MeV. The imaging capabilities of the IBIS are characterized by the coupling of the above quoted source discrimination capability with a very wide field of view (FOV), namely 9$^\circ$ $ \times $ 9$^\circ$ fully coded, 29$^\circ$ $ \times $ 29$^\circ$ partially coded FOV. We present simulations of IBIS observations of the Galactic Center based on the results of the SIGMA Galactic Center survey. They show the capabilities of this instrument in discriminating between different sources while at the same time monitoring a huge FOV. It will be possible to simultaneously take spectra of all of these sources over the FOV even if the sensitivity decreases out of the fully coded area. It is envisaged that a proper exploitation of both the FOV dimension and the source localization capability of the IBIS will be a key factor in maximizing its scientific output. \end{abstract} %SKIP \section*{The IBIS telescope} The IBIS telescope \cite{pg2:Ubertini96} is a coded mask imaging system based on a 1.5 cm thick tungsten mask of 95 $\times$ 95 elements of 1.1 $\times$ 1.1 cm$^2$, designed from a replication of a 53~$\times$~53 MURA basic pattern \cite{pg2:Gottesman89} which gives it a high angular resolution ($\approx$ 12') over a wide field of view (FOV) (9$^\circ$ $ \times $ 9$^\circ$ fully coded, 29$^\circ$ $ \times $ 29$^\circ$ partially coded at zero response). The IBIS detection system is composed of two planes, an upper layer made of 16384 squared CdTe pixels (ISGRI) and a lower layer made of 4096 CsI scintillation bars (PiCsIT). This system enables high sensitivity continuum spectroscopy (E/$\Delta $E$>$10) and a wide spectral range (15 keV -- 10 MeV). \noindent The simulation we performed are for the moment limited to the ISGRI upper layer\cite{pg2:Lebrun95}. The ISGRI pixels are 4$ \times $ 4 mm$^2$, 2~mm thick crystals of Cadmium Telluride, a semiconductor operating at ambient temperature, providing a spectral resolution of about 8$\% $ at room temperature. The 128$ \times $128 = 16384 pixels are arranged in 8 modules separated by dead zones 2 pixel wide needed by the mechanical structures which sustain the detector plane. The sensitivity loss caused by dead zones is not large, however the absence of sensitive elements in the detector plane must be properly taken into account during deconvolution procedures. %Figure 1 --------------------------------------- \begin{figure}[b!] % fig 1 \centerline{\epsfig{file=pg2_1.ps,height=3.5in,width=3.5in}} \vspace{10pt} \caption{\it IBIS/ISGRI Point Source Location Accuracy vs. signal to noise. Values are comprised between triangles} \label{F:pg2:1} \end{figure} %Figure 2 --------------------------------------- \begin{figure}[b!] % fig 2 \centerline{\epsfig{file=pg2_2.ps,height=3.5in,width=3.5in}} \vspace{10pt} \caption{\it 3D image of the IBIS/ISGRI normalized sensitivity over the whole field of view.} \label{F:pg2:2} \end{figure} \section*{Coded Mask Imaging} In the X/gamma-ray domain, source localization can be achieved through the use of coded mask imaging systems \cite{pg2:Caroli87}. These systems represent the most performant way to achieve source localization in this energy domain. In recent years this was demonstrated by the fast localization of X-ray Novae by the SIGMA coded mask telescope \cite{pg2:Vargas97}and by the first localization of the X counterpart of a gamma ray burst thanks to the Wide Field Cameras of the BeppoSAX satellite \cite{pg2:Costa97} . \noindent The incoming radiation is modulated by a mask with opaque and transparent elements and then recorded by a position sensitive detector. In this way it is possible to simultaneously measure source and background thus avoiding the on/off technique usually employed in gamma-ray astronomy which is very sensitive to background variability and observing conditions. The angular resolution of such system is defined by the angle subtended by one hole at the detector which is 12' for IBIS. However point source location accuracy will also depend on the ratio R between mask element size and detector pixel size and will be proportional to the source signal to noise (S/N) ratio. The positional error can be easily computed for the case of integer ratios R as a function of S/N \cite{pg2:Goldwurm95}. ISGRI/IBIS pixels subtend an angle of $\sim$ 4.6' and therefore R is equal to 2.4, we thus expect to reach a position accuracy which is between the values obtained for R=2 and for R=3. These values for a number of S/Ns are reported in Fig. 1. The instrument's extended field of view is divided in two parts, the fully coded field of view (FCFOV), where each source will project a complete (shifted) basic pattern onto the detector plane, and the partially coded FOV (PCFOV) where source flux will be only partially modulated by the mask. Telescope's sensitivity will depend on the amount of modulation and therefore it will be constant in the FCFOV and decreasing with angle from telescope axis in the PCFOV. This is shown in Fig. 2. \noindent Sky images are produced through an algorithm which is basically a balanced correlation between detector image and mask pattern (see \cite{pg2:Goldwurm95}). Reconstructed sky images however are affected by large amount of coding noise in form of ghost peaks and extended modulation due to the sources in the FC or PC FOVs. An iterative source cleaning procedure must therefore been applied in the data analysis to correctly deconvolve sky images. Such procedures have been developed by our group and succesfully used for the data analysis of the SIGMA coded mask telescope images \cite{pg2:Goldwurm95} and now modified and adapted to simulated IBIS images. Indeed the simulations will allow us to test and improve the data analysis techniques. \section*{Imaging simulations of the Galactic Bulge} The Galactic Bulge is the zone of the sky with the highest source density in the X/gamma-ray domain. A typical INTEGRAL Galactic Bulge observation will last about 10$^{5}$~s, i.e. a day. We simulated such an observation in the energy range 40-80~keV assuming sources are at their average brightness level as measured with the SIGMA/GRANAT telescope \cite{pg2:Paul91} during its 7 year long galactic center survey (\cite{pg2:Goldwurm94,pg2:Vargas97}) and adding a contribution also from 4U~1700--37, a hard X-ray high mass binary, at a level about 1/2 lower than the peak flux value detected by SIGMA \cite{pg2:Laurent92}. We then applied sky image reconstruction and cleaning algorithm as described in the previous section. \noindent Results of the simulation (Fig. 3) show the expected imaging performances of the IBIS telescope. All sources including the faint Terzan~1 ($\approx$~8~mCrabs) are clearly detected at more then 5 sigma level with the brighter (the ``microquasar'' 1E~1740.7-2942) at more than 40 sigma. The sources are well separated even in the very center of the galaxy where source confusion problems may arise. It can be seen that 4U~1700--37 is also easily detected by the IBIS instrument, which proves the possibility of taking a high significance spectrum even if the source is at more than 11$^{\circ}$ from pointing axis. The possibility to monitor such huge field gives to IBIS a very high potential for search and high precision ($<$~1') localization of high-energy brigth transient sources like X-ray novae and gamma-ray bursts. Moreover thanks to the wide field and the galatic plane survey program, it will be possible to detect serendipitous, faint sources like Ti$^{44}$ lines from hidden supernovae in the galactic plane. %Figure 3 --------------------------------------- \begin{figure}[b!] % fig 3 \centerline{\epsfig{file=pg2_3.ps,height=5.8in,width=6in}} \vspace{10pt} \caption{\it 5-level contour image of the complete ISGRI FOV of the Galactic Center in the 40-80 keV band. Units on axis are sky pixels (about 4.6 arcminutes see text). Note the clear detection of the 4U1700-377 flare at more than 11 $^{\circ}$ from the pointing direction} \label{F:pg2:3} \end{figure} \begin{references} \bibitem{pg2:Caroli87}Caroli E. et al., {\it Space Sci. Rev..} {\bf 45}, 349 (1987) \bibitem{pg2:Costa97} Costa E. et al., {\it Nat.} {\bf 387}, 743 (1997) \bibitem{pg2:Gottesman89}Gottesman S. R. and Fenimore E., {\it Applied Optics.} {\bf 28}, 20 (1989). \bibitem{pg2:Laurent92}Laurent, P. et al., {\it A.\&A.} {\bf 260}, 237 (1992) \bibitem{pg2:Lebrun95}Lebrun F. et al., Proc. of "Imaging in High Energy Astronomy", Eds. L. Bassani and G. DiCocco, Kluwer Acad., The Netherlands (1995) \bibitem{pg2:Goldwurm95} Goldwurm A., {\it Exper. Astron.} {\bf 6} 9 (1995) \bibitem{pg2:Goldwurm94} Goldwurm A. et al., {\it Nature}, {\bf 371} 589 (1994) \bibitem{pg2:Paul91} Paul J. et al., {\it Adv. Space Res.}, {\bf 11}(8), 289 (1991) \bibitem{pg2:Vargas97} Vargas M. et al., Proc. of 2nd INTEGRAL Workshop, {\it ESA SP}-382, 129 (1997) \bibitem{pg2:Ubertini96} Ubertini P. et al., {\it SPIE}, {\bf 2806}, 24 (1996) \end{references} \end{document} ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message ----- ----- End Included Message -----