1. MARS – CheObs ed.* A Flexible Software Framework for Future Cherenkov Telescopes Thomas Bretz and Daniela Dorner presented by Pavel Binko * Modular Analysis and Reconstruction Software – Cherenkov Observatory edition 1 11th ICATPP Conference, Como

2. Outline ● Introduction: – Imaging Air Cherenkov Technique – Analysis Principles – Existing and future Cherenkov Telescopes ● DWARF Project ● Mars – CheObs ed. – Framework – Simulation ● Outlook 2 11th ICATPP Conference, Como

3. Imaging Air Cherenkov Technique ~ 10 km Particle: gamma or cosmic ray 3a 11th ICATPP Conference, Como

4. Imaging Air Cherenkov Technique ~ 10 km Shower of secondary particles 3b 11th ICATPP Conference, Como

5. Imaging Air Cherenkov Technique ~ 10 km ~1o~1o Cherenkov light ~ 120 m Emission of Cherenkov light 3c Light flashes of few nano-seconds 11th ICATPP Conference, Como

6. Imaging Air Cherenkov Technique ~ 10 km ~1o~1o Cherenkov light ~ 120 m 1 Telescope => Mono Observations Image in the camera 3d 11th ICATPP Conference, Como

7. Imaging Air Cherenkov Technique Array of Telescopes => Stereo Observations 3e 11th ICATPP Conference, Como

8. Analysis Principles Data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Calibration Relative/Absolute calibration Treatment of defect pixels 4a 11th ICATPP Conference, Como

9. Analysis Principles Data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Calibration Relative/Absolute calibration Treatment of defect pixels 4b 11th ICATPP Conference, Como

10. Analysis Principles Data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Calibration Relative/Absolute calibration Treatment of defect pixels Statistical analysis of the light distribution (size, mean, rms) 4c 11th ICATPP Conference, Como

11. Different primary => different shower morphology hadronic background muon ring gamma:background 1:1000 gamma candidate 11th ICATPP Conference, Como 4d

12. Analysis Principles Data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Calibration Relative/Absolute calibration Treatment of defect pixels 4e Muons used for absolute light calibration 11th ICATPP Conference, Como

13. Analysis Principles Data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Calibration Relative/Absolute calibration Treatment of defect pixels 4f 11th ICATPP Conference, Como

14. Analysis Principles Data Cuts Training with other data Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time 4g 11th ICATPP Conference, Como

15. Background images 11th ICATPP Conference, Como 4h

16. Real data Simulated muons Simulated gammas Distribution of image parameter Background images 11th ICATPP Conference, Como 4i

17. Real data Simulated muons Simulated gammas Distribution of image parameter => cut Background images Discriminate Gammas from Backgound 11th ICATPP Conference, Como 4j

18. Analysis Principles Data Cuts Training with other data Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin Calculation of the significance of the observation Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time 4k 11th ICATPP Conference, Como

19. alpha theta² Reconstruction of the shower origin Albert et al. 2006 4l Dorner 2008 θ² θ² 11th ICATPP Conference, Como

20. 4 Analysis Principles Data Cuts Training with other data Calibration Relative/Absolute calibration Treatment of defect pixels θ²θ² Dorner 2008 Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Events from the source: θ² or α values close to 0 11th ICATPP Conference, Como 4m

21. Analysis Principles Data Monte Carlo Simulations Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Cuts Training with other data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Simulate events with same conditions like data: psf, atmosphere, night sky background Same analysis chain like data => image parameters + energy of primary particle 4n 11th ICATPP Conference, Como

22. Energy of the primary particle _ shower size Other influences: impact parameter, zenith angle,... 4o Same energy Different impact => different distance => different size Example from DWARF simulation: Simulated gamma events for different impact parameters Energy Reconstruction using simulated events => reconstruct energy from parameters 11th ICATPP Conference, Como

23. Analysis Principles Monte Carlo Simulations Data Energy reconstruction Training of a Random Forest or Parametrization Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Cuts Training with other data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time 4p 11th ICATPP Conference, Como

24. Analysis Principles Monte Carlo Simulations Data Spectrum Energy reconstruction Binning Effective collection area Spillover correction Spectrum Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Cuts Training with other data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Energy reconstruction Training of a Random Forest or Parametrization 4q 11th ICATPP Conference, Como

25. Analysis Principles Monte Carlo Simulations Data Energy reconstruction Training of a Random Forest or Parametrization Calibration Relative/Absolute calibration Treatment of defect pixels Background-supression Quality-cuts/Background-supression Reconstruction of the shower origin => θ 2 Calculation of the significance of the observation Cuts Training with other data Image-Parameter Software-Trigger Image cleaning Calculation of parameters Muon analysis Calculation of effective observation time Dorner 2008 4r 11th ICATPP Conference, Como

26. H.E.S.S. VERITA S CANGARO O MAGI C Current Generation of Cherenkov Telescopes 5 11th ICATPP Conference, Como

27. Future Cherenkov Telescopes ● Discovery of many new sources in wider energy range with improved sensitivity: AGIS, CTA Advanced Gamma-ray Imaging System Cherenkov Telescope Array 6a 11th ICATPP Conference, Como

28. Future Cherenkov Telescopes ● Discovery of many new sources in wider energy range with improved sensitivity: AGIS, CTA ● Longterm monitoring of bright AGN: DWARF photomontage Advanced Gamma-ray Imaging System Dedicated multiWavelength AGN Research Facility Cherenkov Telescope Array 6b 11th ICATPP Conference, Como

29. DWARF Project Active Galactic Nuclei Extreme variablity on time scales from minutes to years possible explanation e.g. binary black holes Correlations between different wavelengths => better understanding of acceleration mechanisms Especially correlations with neutrinos (IceCube), clear hadronic signature 7a 11th ICATPP Conference, Como

30. DWARF Project Long-term observations mandatory! Active Galactic Nuclei Extreme variablity on time scales from minutes to years possible explanation e.g. binary black holes Correlations between different wavelengths => better understanding of acceleration mechanisms Especially correlations with neutrinos (IceCube), clear hadronic signature 7b 11th ICATPP Conference, Como

31. DWARF Project The idea Affordable monitoring telescope observing only a few target objects: Mkn421, Mkn501, 1ES 1959+650, 1H 1426+480, PKS 2155-304, 1ES 2344+514 with large duty cycles Refurbish one of the former HEGRA telescopes switched off since ~2002 CT3 still at La Palma HEGRA CT3 at La Palma 8a 11th ICATPP Conference, Como

32. DWARF Project HEGRA CT3 at La PalmaDWAR F 8b 11th ICATPP Conference, Como

33. DWARF Project Upgrade telescope: Will be operated remotely 40 mirrors from HEGRA CT1 hexagonal after refurbishment: slightly larger mirror area, improved reflectivity New type camera Semi-conducting photosensors (G-APDs) Compared to PMTs: better photon detection efficiency, robustness Use of solid cones under investigation Design ready end of the year, camera finished in 2010 DWAR F 8c 11th ICATPP Conference, Como

34. DWARF Project Mirro r Prototype camera Prototype: First shower images successfully recorded Extensive lab-investigations ongoing Anderhub et al. submitted to JINST 9 11th ICATPP Conference, Como

35. MARS – CheObs ed. Bretz, AIP Conf.Proc. 745, 2005 General functionalities Analysis methods IACT algorithms 10a Detector setup 11th ICATPP Conference, Como Software: Layer Architecture => Flexibility => usable also for other experiments

36. reflector geometries camera geometries 10b MAGIC I camera MAGIC reflector DWARF reflector Detector Setup two possible DWARF camera layouts 11th ICATPP Conference, Como

37. MARS – CheObs ed. Bretz, AIP Conf.Proc. 745, 2005 Software: Layer Architecture => Flexibility => usable also for other experiments General functionalities Analysis methods IACT algorithms Detector setup Telescope specific 10c 11th ICATPP Conference, Como

38. MARS – CheObs ed. Bretz, AIP Conf.Proc. 745, 2005 Plugin-Concept: Easy enhancement, e.g. new algorithms 11 11th ICATPP Conference, Como

39. MARS – CheObs ed. Very flexible logging stream Online display (tracing) Histogram classes Debugging (runtime, calls, etc) Generalized resource-files Easy to implement Standardized Very powerful Complete(!) setup is stored in output file (reproducibility) All algorithms telescope- / data- independent Generalized I/O (for all data), e.g. every output is easily implemented reading of different data format is very easy User Friendliness 12 11th ICATPP Conference, Como

40. MARS – CheObs ed. cron script Software Analysis Status Data Quality Results Processing Status/Control Data Quality / Data Check High Level Science Data Compilation of own analysis from own data Data, MC, Input Output, Results Package includes automatic processing Database queuing system Storage 13 11th ICATPP Conference, Como

41. Simulations for DWARF MARS – CheObs ed.: Simulations for DWARF: ceres* Simulation studies for optimizing the camera properties Comparison of several trigger types More than 50 triggers simulated Sum trigger performs clearly best * Camera Electronics and REflector Simulation 14a 11th ICATPP Conference, Como

42. Simulations for DWARF ~400GeV Flux, Slope=-2 PRELIMINARY ! MARS – CheObs ed.: Simulations for DWARF: ceres* Simulation studies for optimizing the camera properties Comparison of several trigger types More than 50 triggers simulated Sum trigger performs clearly best All simulations show promising results with an energy threshold (flux peak energy) around 400GeV for reconstructed images 14b * Camera Electronics and REflector Simulation 11th ICATPP Conference, Como

43. Conclusions ● MARS – CheObs ed.: – Flexible and userfriendly software package not only for Cherenkov telescopes – New: includes package for simulation of IACTs – Used for DWARF, MAGIC and CTA ● DWARF: – Successfully recorded first shower images with prototype camera FACT – Simulations very promising for physics goals 15 11th ICATPP Conference, Como

44. Outlook ● MARS – CheObs ed.: – Implementation of stereo analysis ● Dedicated Worldwide AGN Research Facility – Network of small Cherenkov telescopes around the globe – 24/7 monitoring – Starting point: collaboration with Whipple 10 m 16 11th ICATPP Conference, Como