ISSI meeting Mar 2011.pptx - User Web Pages - Monash University

The Multi-INstrument BurstARchive – MINBARDuncan GallowayMonash UniversityJean in ‘t ZandSRONJérôme Chenevez &Søren BrandtDTU SpaceLaurens KeekU. MinnesotaISSI Meeting #1, Mar 2011, BernDuncan.Galloway@monash.eduhttp://users.monash.edu.au/~dgallow

Large thermonuclear burst samples• Catalogues from large samples of burstsanalysed with uniform data analysisprocedures have proven a valuable resourcefor investigations of burst energetics,thermonuclear processes, and burstoscillations– Van Paradijs, Penninx & Lewin 1988 (althoughcombined data from bursters with different accretedcompositions)– Cornelisse et al. 2003 study of BeppoSAX data– Muno et al. 2001, 2002a,b 2003, 2004 studies ofburst oscillations– RXTE burst catalog (Galloway et al. 2008)Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

The RXTE burst catalog• An almost decade-long effort led to the 2008release of the first multi-source burstcatalogue (Galloway et al. 2008; hereafter G08), thatincluded analysis results for almost 1200events, from 40 burst sources observed byRXTE.• Amongst other things,analysis of this dataestablished the needto distinguishbetween differentsource classes (mixedH/He and ~pure Heaccretors)Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

Some improvements can be made• This catalogue has become a useful referencefor observers and theorists, but is notwithout its shortcomings:– Reliance on RXTE pointed observations resulted inlittle exposure for many sources. Wide-fieldinstruments (e.g. BeppoSAX) gave much moreexposure, particularly in the Galactic centre– RXTE data consequently had a poor detection ratefor rare events like intermediate-duration bursts– RXTE has no imaging capability so localisingbursts, particularly in crowded regions such as theGalactic centre, is a challenge• RXTE has continued observations of burstsources, so there are additional bursts innew public data -> need to keep updatingGalloway, “The Multi-INstrument Burst ARchive – MINBAR”

MINBAR seeks to address these issues• Assembly of the Multi-INstrument BurstARchive (MINBAR) is under way. The completecatalogue will have the following additionalbenefits over G08:– updated analysis procedures, bettersuited to comparisons of burstsbetween different instruments– analysis with latest RXTE responsematrices, and improved treatment ofcolumn densities– first public release of re-analysedBeppoSAX/WFC burst data– first public release of companionRXTE observation database (notreleased as part of the G08catalogue)Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

Additional minor benefits• Better measurement of burst rates etc.thanks to “interleaving” of burstobservations• Practically this is rather limited, due tolow duty cycles for individual instruments.For example, the overlap between theBeppoSAX and G08 samples (2200/1200) burstswas such that only 15 bursts were observedby both instruments• But such overlap can be critical forrecurrence time studies (e.g. KS 1731-26, &4U 1728-34 with Chandra)Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

MINBAR current version (0.51)• Total of 4175 bursts from 72 sources (fromRXTE PCA and BeppoSAX WFC data)• Some verification has already taken place,but more to do• Currently (at Monash) working on identifyinglong H/He bursts and matching to modelresults (with Lampe, Keek & Heger)• Also produced a preliminary list of longbursts from MINBAR and the literature forour deliverable from this meeting (ask mefor a copy!)Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

Galloway, “The Multi-INstrument Burst ARchive – MINBAR”

Results so far• Pre-release versions including RXTE andBeppoSAX data have already been the subjectof detailed analysis• Keek et al. (2010) presented an analysis of136 bursts from the combined RXTE andBeppoSAX data with recurrence times

Fig. 2. Spectral fits to the PCA bursts from 4U 0614+09 and 4U 1722-30 during time intervals when the absorbed black body plus reflection modelfails to fit the data (see Fig. B.1 for example time intervals when the black body plus reflection model provides a good fit to these bursts). LeftResults #2• in ‘t Zand & Weinberg (2010) presentedanalysis of “superexpansion” bursts whichshowed significant spectral fit residualscompared to a blackbody model• Suggested to arise from absorption edgesfrom material ejected during the RE4 J. J.M. in’tZandandN. N.Weinberg: Evidence forheavy-element ashes in superexpansion X-ray bursts10 44U 0614+09 at 1.7 s / Black body + reflection fit10 4 4U 0614+09 at 2.5 s / Black body + reflection fit4U 1722−30 / Black body + reflection fit[phot s −1 keV −1 ]10001001000100100010010Data/Model21.510.55 102021.510.55 10201.41.210.85 10204U 0614+09 at 1.7 s / Fit with edge included4U 0614+09 at 2.5 s / Fit with edge included4U 1722−30 / Fit with edge includedData/Model0.5 1 1.5 25 1020Energy [keV]Galloway, “The Multi-INstrument Burst ARchive – MINBAR”0.5 1 1.5 25 1020Energy [keV]0.8 1 1.2 1.45 1020Energy [keV]

Results #3A&A 525, A111 (2011)• in ‘t Zand &cpresented analysis ofanother strong REburst that showedremarkable variabilitylate in the burst• Possible this arisesfrom the interactionof the expanded shelland the accretion diskGalloway, “The Multi-INstrument Burst ARchive – MINBAR”Rate > 4 keV [c/s]Rate < 4 keV [c/s]0 5000 10 4 1.5×10 0 4 2000 4000 6000Hardness0 50 100 200 3000 1 2 3Time since burst onset [s]0 0.02 0.04 0.06Fig. 1. PCA lightcurves of long burst from 2S 0918-549 in two bandpassesand over the two active PCUs (top two panels) andthehardnessratio between both (third panel). The time resolution is 0.1 s for timesearlier than 1 s, and 1.0 s for later times. The inset in the first panelshows the full-bandpass lightcurve for the first 60 ms at a time resolu-Rate [c/s]0 5000 10 4 1.5×10 4Leahy power10 100 1000 10 41200.

Result #4• Combined study of the bursting behaviour ofthe182newJ. Cheneveztransientet al.IGR J17473-2721IGR J17473–2721 outburst 2008 and burst timesRXTE #32Bolometric fluxRXTE/ASM (1.5–12 keV)Swift/BAT (15–50 keV)INTEGRAL/JEMXRXTE/PCAFlux (10 –9 erg/cm 2 /s)0 5 10 15JEMX #1SuperAGILESwiftRXTE #35RXTE #33+34RXTE #36RXTE #41RXTE #42500 550 600 650 700 750Time (MJD – 54000)Figure 1. Broad-band X-ray light curves for IGR J17473−2721 during the 2008 outburst, combining all available RXTE and Swift/BAT 1-day averaged data.Galloway, “The Multi-INstrument Burst ARchive – MINBAR”The bolometric fluxes are derived from RXTE/(PCA+HEXTE) fits (see the text) and have been connected for clarity. Every PCA observation is representedby a black dot. The observed X-ray bursts are represented by stars, with circles for the four PRE ones, and those indicated by names delineate the transitionsbetween different phases of burst behaviour (see Sections 3.3 and 4.1).

Issues• Still have some work to do to get all thedata we need (e.g. lightcurves) for the datain hand• Switch to XSpec 12 revealed an issue withthe spectral fits for RXTE bins with zerocounts. Fixed with different choice ofweighting for chi^2, but some exceptionsnoted• No JEM-X bursts yet• No strategy yet (nor personnel) for addingbursts from other instruments. We arepersonnel limited in generalGalloway, “The Multi-INstrument Burst ARchive – MINBAR”

The future• I’ll spend two weeks in Denmark working onJEM-X burst analysis, and the focus thisyear will be getting those bursts in thecatalog in some form• Hoping for some contributions from the ESTECgroup (Sanchez, Kuulkers)• Likely spend another 2-3 weeks in June• Once the catalog is complete (with burstsfrom all three instruments) we will have a“proprietary” period of not less than 6months and then release the catalog publicly• In the meantime, science work will continueon the preliminary versionsGalloway, “The Multi-INstrument Burst ARchive – MINBAR”