CUC Apr. 5, 2006HLM — XMM/ChandraCross-Calibration between XMM-Newton and ChandraHerman L. MarshallChandra X-ray Center, MIT Kavli InstituteOverviewXMM/Chandra Cross-calibration telecons:continuing monthly meetingsformed web pages comparing fitsBegan effort to cross-cal formallyGoal: achieve χ2 = 1 for all cal sourcesprototyping methodology continuesAttending meetings for joint cal discussionsnext EPIC cal meeting in May (Munich)multi-observatory cross-cal meeting in June (Reykjavik)Implementing cross-cal with Suzaku in MayFrom http://space.mit.edu/ASC/calib/crosscal/XMM/Chandra Blazar Observations to DateFrom XMM team: XMM-SOC-CAL-TN-0052 (Stuhlinger et al. 2006)Sample XMM-Chandra Comparison (XMM side)From XMM team: XMM-SOC-CAL-TN-0052 (Stuhlinger et al. 2006)Comparing XMM and ChandraXMM web pagesFit results internal for nowdata page: http://xmm.esac.esa.int/~xmmdoc/EPIC_CROSSCAL/sat_cross_cal.phpXMM report: XMM-SOC-CAL-TN-0052 (Stuhlinger et al. 2006)CXC web page: http://space.mit.edu/ASC/calib/crosscal/XMM whiteboard used for some reportsResults agree generally but not within errors From http://xmm.esac.esa.int/~xmmdoc/EPIC_CROSSCAL/cross_cal_data.phpComparing indices is difficultSample XMM-Chandra Comparison (XMM side)From http://space.mit.edu/ASC/calib/crosscal/Sample XMM-Chandra Comparison (HETGS side)Caveats: XMM-Chandra Comparison (HETGS side)Variability: affects one observation of PKS 2155-304 (but not 1H 1426+428 and 3C 273)Different models used by ESAC and CXCbroken PL vs. smooth PLHETGS fits without 2PLResiduals: compare for complete storyBow-tie: models extrapolated to low SNRHave not achieved χ2 = 1 for bright sourcesFormalizing Cross-CalibrationGoal: achieve χ2/ν = 1 for all cal sourcesPremise: Users want χ2/ν = 1 if model fitsLocal adjustments, Ajn g(x; βj) -- ---> overdetermined system, reducing χ2Ajn depend on instrument n, βj fixedCompare instruments by comparing AjnMethod proposed: Lagrange multipliersMinimize where Formalizing Cross-CalibrationTest casesMk 421: bright, 100 ksXTE J1118+480: bright 24 ksg(x) = Gaussian, σ = 2Å, 2Å apart, starting at 1ÅResultsOK amplitudes (figure)min χ2/νseparately: 2.20, 1.14jointly: 2.62, 1.48ConclusionsNeed more Gaussians or different basis functions or adaptive functions ...Work in ProgressJoint web page populationCoordinate modeling between projectsProcess all Chandra cal data uniformlyWill add analysis of 1E 0102-72Developing joint analysis methodologyPrimary goals may not be achievable soonMust define secondary goalsSetting up cross-cal with XMM and SuzakuTarget: PKS 2155-304Week of May 2, 2006Will use LETG/HRCChandra grating observations of PKS 2155Aug ‘06 — independent of XMM, SuzakuTo be used for Chandra internal cross-calOverviewFormed web pages comparing fitsGeneral but not statistical agreementResults need revision and analysisPrototyping formal cross-cal methodologyAttending meetings for joint cal discussionsImproves approach to data handlingRefines analysis and methodologyImplementing cross-cal with Suzaku in May