- Slides: 13
Extragalactic Survey with MAXI and First MAXI/GSC Catalog Yoshihiro Ueda Kazuo Hiroi, Naoki Isobe, Masaaki Hayashida (Kyoto University) Mutsumi Sugizaki (RIKEN) and the MAXI team
Introduction n MAXI achieves the best sensitivity in the 2 -10 ke. V band as an all sky survey mission after HEAO-1 and RXTE slew survey. n n Deeper than ROSAT All Sky Survey for log NH >22 Deeper than Swift/BAT or INTEGRAL for log NH <23 with “soft” spectra (e. g. , narrow line Seyfert 1 s) Unique opportunity to detect highly variable objects We report first results of the MAXI/GSC catalog at high Galactic latitudes (|b|> 10 deg) see Hiroi+ poster n n 7 months data of 8 cameras from 2009 Sep to 2010 Mar (operated with HV=1650 V) Energy band: 4 -10 ke. V (best calibrated)
Revised estimate of MAXI sensitivity Comparison with our pre-launch estimate (YU+ 2009) n Observation efficiency ~40%, due to the HV turn-off at high latitudes and the reduced number of cameras (12→ 8) n Slight update of the non X-ray background rate. Below 10 ke. V it is consistent with previous missions (Sugizaki’s talk) n Expected 5σ sensitivity in the 2 -10 ke. V (4 -10 ke. V) band ~1. 3 (2. 0) m. Crab ~0. 5 (0. 8) m. Crab ~0. 2 (0. 3) m. Crab from 1 month operation from 7 months operation from 3 years operation cf. confusion limit (1 source per 10 beams) ~ 0. 2 m. Crab
Construction of background model n The non X-ray backgroud of MAXI/GSC shows very complex behavior, largely affected by the time dependent environment of the ISS (eg, solar paddle, shuttle docking). We made the database of the background for each camera as a function of epoch, based on onboard calibration. Short variability (
Construction of background model Example of blank sky in an accumulated MAXI image: data vs background model (NXB+CXB) Projection n Dec R. A.
Step 1: Search for Source Candidates Analysis is made for each projected image of 10 deg x 10 deg n Smooth with a circle of r=1 degree (ie, “sliding circle”) n By subtracting the background from the real data, we search for excess signals whose significance is defined as residual/sqrt (real data). n From the sky of |b| > 10 degree, we picked up total ~300 source candidates with significance >4. 5 σ. 10 degree smoothing circle
Step 2: Source Characterization n Fitting to image in photon counts by maximum-likelihood algorithm, utilizing a model consisting of point spread functions and background. PSFs are produced by the MAXI simulator, where detailed observing conditions are fully considered. n n n Determine flux and position with their statistical errors for each source. Detection significance is defined as σD= (best-fit flux) / (1σ error) It is essentially same as “likelihood”. Finally, human inspection to remove obviously fake detections (e. g. , around Sco X-1). Projection PSFs
Results (|b|>10, 4 -10 ke. V) ( 227 (136) sources above 5σ (7σ) Preliminary >7σ detection
Sensitivity Curve and log N – log S n n n Survey area as a function of limiting flux is estimated from the all-sky map of background and of [effective area ×exposure]. The expected sensitivity (~0. 8 m. Crab at 5σ) confirmed. Extragalactic log N – log S well matches with the HEAO 1 A 2 result (Piccinotti+ 1982) 0. 8 m. Crab 5σ 7σ HEAO 1
Position Accuracy n n n Angular separation between MAXI positions and most likely counterparts, given as a function of significance Systematic positional error is ~0. 05 deg (90% radius) Statistical (and related) error increases as ∝ 1/σD 7 -10σ Averaged seperation (not 90% radius !!) 10 -30σ > 30σ
Cross Correlation with other Catalogs MAXI total (>7σ) Swift/BAT 58 -months Fermi/LAT 1 year RASS (above 0. 3 count/s) at least one 136 69 10 75 95 Preliminary ID status Galactic/MC objects (red) 23 Clusters of galaxies (orange) 29 AGN and un. ID (blue) 37+47
AGN: flux comparison with HEAO 1 n About half of the 31 AGNs in the HEAO 1 A 2 catalog (Piccinotti+ 1982) are not detected with MAXI above 7σ (with a better sensitivity than HEAO-1): the list of “brightest AGNs” has changed dramatically from 30 years ago. n AGNs commonly detected with MAXI and HEAO 1 show evidence for time variability in most cases.
Summary and Future Plan n The 7 -months MAXI/GSC catalog at |b| > 10 contains >220 sources above 5σ with a limiting sensitivity of 0. 8 m. Crab in the 4 -10 ke. V band. n n Predicted sensitivities under the actual observation conditions are confirmed. More sources expected with longer integration and inclusion of 2 -4 ke. V data. Source identification work is on-going. Swift/XRT will help a lot for unidentified objects. Better understanding of the NXB would be required. We plan to perform source detection in time sliced data to search for “transient” objects. This will become the unique feature of the MAXI survey.