Data Challenge II

1
Data Challenge II
Logo by Stefano Ciprini
2
Data Challenge 2

• This is the beginning of DC2, but getting to the
  beginning has been the result of a lot of hard
  work by many people over many months.
• Work was organised by a steering committee Jean
  Ballet, Toby Burnett, Rob Cameron, Valerie
  Connaughton, Seth Digel, Richard Dubois,
  Francesco Longo, Julie McEnery, Steve Ritz and
  Tracy Usher
• But the actual work has been done by a much
  larger number of people.
• Thanks to everyone who helped!

3
DC2 What is it?

• 55 day simulation of the high energy gamma-ray
  sky
• Full detector simulation
• Detailed modeling of astrophysical sources
• Include updated background model including orbit
  dependent effects.
• Goal was to make the DC2 data as realistic as
  possible
• Science analysis software and definition of the
  data products have been made mature

4
Reconstruction and simulation geometry

• Updated and checked detector geometry
• Introduced some realistic elements to the
  detector model dead strips etc. (we did this
  before we had a full LAT to get real
  measurements)
• Revamped TkrRecon and CalRecon
• Many new variables added to support background
  rejection and event classification analysis.

5
Background Model, Deadtime, orbit model

• Updated background model including orbit
  dependent effects
• Included effect of deadtime (fractional deadtime
  depends on the event rate after the throttle is
  applied).
• Updated orbit and attitude profile including
  realistic slew rates and yaw flips.

Plot of livetime fraction vs geomagnetic latitude
(illustrates the effect or orbit dependent
trigger rates)
6
Event Classification and background rejection

• Several iterations (by several people) of
  classification tree analyses to identify
  well-reconstructed gamma-ray events.
• This converged into an analysis by Bill Atwood,
  the final result is a small number of high-level
  variables which can be used as knobs to tune
  performance for PSF, Energy resolution and
  background rejection
• CTBGAM (Energy Resolution)
• CTBCORE (PSF)
• CTBBestEnergyProb (Energy resolution)

7
Analysis Classes

• The infinite range of possibilities with the
  knobs provided by the classification analysis
  have been distilled into two analysis classes.
• Class A These are a set of cuts that provide
  good background rejection, angular resolution and
  energy resolution.
• All These are a looser set of cuts which provide
  additional effective area and may be suitable for
  analyses where background contamination less
  important.
• Use gtselect to create a filtered event file
  creating events for the analysis class of
  interest.
• Select the relevant response functions in the
  science tools (DC2_A for classA and DC2 for all).
  The selections made in the data file must match
  the choice of IRF used in the analysis.

8
Preliminary Catalog Analysis

• A preliminary catalog analysis has been performed
  (by Jean Ballet and colleagues)
• This analysis was run autonomously on the catalog
  pipeline with minimal human interaction (so not a
  final polished product but an excellent
  demonstration of our current abilities)
• You can use this as a starting point to identify
  interesting looking sources to study in greater
  detail.
• On Monday there will be an updated version which
  will have correct test statistic and
  Significance.

9
LAT Data Products

• Root
• MeritTuple (event based summary information)
• pointing_history (spacecraft location and
  attitude)
• Recon and Digi trees (detailed low level
  information)
• Fits
• FT1/photon data (event based summary information
  for events which belong to the all analysis
  class)
• FT2/pointing data (spacecraft location and
  attitude)
• Higher level data (also fits)
• Livetimecubes intermediate data product which
  speeds up exposure calculations.
• Pulsar Ephemerides
• Model of the Galactic diffuse emission
• Preliminary LAT point source catalog

10
GBM simulations and data

• Simulated GBM data has been generated for a
  population of bursts.
• This includes all the burst data products and
  also includes a GCN-like notice to provide
  summary GBM information.
• This allows us to more fully explore the
  possibilities of GRB studies in the GLAST era.

11
Data data data data.

• We generated lots and lots of data.
• The efficient running of many analysis efforts
  depended on them being fed promptly with
  simulated data.
• The system worked! We ended up generating very
  much more data than we first estimated in a much
  shorter time.
• The system to track data and allow reprocessing
  worked well (and there was a lot to track)
• Most of all, there was a human in the loop that
  made datasets appear seemingly effortlessly.

Many, many thanks to Tom Glanzman!
12
The DC2 Sky

• DC2 sky in galactic coordinates

Plot by Seth Digel
13
DC2 Sky Galactic Center region
14
Contents/features of the data

• The DC2 data contains
• Steady point sources
• Variable sources
• Flaring
• Bursting
• Periodic
• Spatially extended sources over several size
  scales
• Many of the DC2 sources are located on top of
  known astrophysical sources (certainly most of
  the bright ones).

15
DC2 Goals, requirements and purpose

• 55 days of LAT data provide a deeper view of the
  high energy gamma-ray sky than has previously
  been achieved.
• Results from previous gamma-ray missions provide,
  at best, an incomplete guide to the DC2 sky.
• Part of the challenge of DC2 will be to figure
  out what was included in the sky model.
• DC2 data has a fairly realistic level of detail
  which will support a wide variety of both science
  and instrument performance studies.
• Exercise the science tools but dont feel
  restricted to them
• Improve the documentation and analysis software
  from user feedback.

16
Produce LAT point source catalog

• This will be an update of the catalog produced at
  the beginning of DC2.
• See https//confluence.slac.stanford.edu/display/S
  T/LATSourceCatalogContents for a description
  (and discussion) of the contents of the LAT
  point-source catalog.
• Requirement Spectral index and flux (with
  associated uncertainties), location with 68 and
  95 confidence ranges, flux in discrete energy
  bands.
• Goal Variability index, flux history, peak flux,
  measure of whether a source is extended.

17
Develop and test source detection algorithms

• Source detection is a necessary step in producing
  the LAT point source catalog. Several methods to
  do this have already been developed (see talks on
  Thursday), more may be developed during the
  course of DC2
• Requirement That these algorithms are tested and
  compared with one another in a systematic way
  using the DC2 data.

18
Pulsars (1)

• There are many simulated pulsars in the DC2 data.
  We provide an ephemeris for most of the
  radio-loud pulsars. These ephemerides are
  mostly (but not always) valid during the DC2 time
  period, i.e. in some cases the ephemeris can be
  only considered as a guide.
• Requirement Determine the gamma-ray lightcurves
  for at least 6 pulsars which have an exact
  ephemeris.
• This can be done using the science tools (gtbary,
  gtpphase)
• Requirement Determine timing properties of
  pulsars and produce gamma-ray lightcurves for at
  least one pulsar with an approximate ephemeris.
• This can be done using the science tools (gtbary,
  gtpsearch)
• Goal Determine lightcurves for more of the
  fainter pulsars in the DC2 data.
• Goal Phase resolved spectra for bright (EGRET)
  pulsars.

19
Pulsars (2)

• Not all of the pulsars will be included in the
  ephemerides.
• Goal blind periodicity searches on candidate DC2
  pulsars
• Goal Pulsar population studies the ratio of
  radio-loud to radio-quiet pulsars is sensitive to
  the emission mechanism for producing the
  gamma-rays. It will be possible to study this in
  the DC2 data by comparing the number of pulsars
  with ephemerides to those without.

20
Extended sources

• Several source types may produce objects which
  can be resolved by the LAT as extended sources
  SNRs, Galaxy Clusters, Dark matter
• Goal To identify extended sources in the DC2
  data (there are some)
• Goal Perform spatially resolved spectroscopy.

21
Variable sources

• Many sources in the DC2 sky are variable.
• Requirement Produce lightcurves for at least 20
  bright sources (from the data release plan, these
  are the sources we will release high level data
  from in year 1)
• Goal look at lightcurves for many more sources
• There are several variable sources in the DC2 sky
  which are not AGN or GRB. Some of these may have
  very distinct variability properties.
• Goal To find and study variable sources that
  might not be blazars (i.e. the AGN folk do not
  get to have all the fun)

22
Spectral Studies

• Sources in the DC2 data may have spectra that are
  more complicated than simple power-laws. This
  opens up many analysis possibilities
• Goal Study spectra of pulsars to determine the
  shape of spectral cutoffs
• Goal EBL attenuation studies (redshift dependent
  cutoffs)
• Goal Search for spectral signatures of dark
  matter

23
Gamma-ray bursts (1)

• Gamma-ray bursts are included in the DC2 sky.
  This includes both LAT and GBM data. There are
  also GCN-like notices which provide summary
  information about the GBM flux and location
  (including statistical uncertainties).
• Requirement Perform joint spectral fits of at
  least one burst using both LAT and GBM data.
  (gtbin, rspgen, xspec)
• Requirement Temporal studies/comparison of at
  least one GRB
• Requirement Produce preliminary GRB catalog,
  this should include GBM LAT properties (goal
  include LAT upper limits for GRB with no LAT
  detection).

24
Gamma-Ray Bursts (2)

• Goal Search for LAT only GRB
• Goal Search for additional high energy
  components and/or afterglows
• Goal Compare the LAT and GRB locations and
  quoted statistical uncertainties to study the
  systematic GBM localisation uncertainty.

25
Other sources

• Requirement Identify at least one source that is
  not a pulsar, AGN or GRB (there are some that can
  be unambiguously identified from the gamma-ray
  data)
• This could be done based on location
• Or location combined with variability properties
• Or by a discovery of periodic emission

26
Diffuse sources

• There are two major components of diffuse
  emission, extragalactic and galactic. Studies of
  these are likely to be quite sensitive to the
  presence of residual background.
• Goal Study flux, spectra and spatial
  distribution of the galactic diffuse and compare
  with the diffuse model provided for source
  analysis.
• Goal Study flux and spectral properties of the
  extragalactic background. This will include a
  study of the effect of residual background,
  contribution from galactic diffuse and resolving
  the point sources.

27
Quicklook and transient release tasks

• During the first year of LAT operations we will
  be releasing data to the public on 20 selected
  sources. The DC2 data can be used to explore how
  this will work and what information is useful.
• Requirement define the format of the summary
  data that will be released for these objects
  (i.e. what variables etc) and pull the analysis
  results into this format.
• Goal Search for and produce time resolved
  summary data for any source that exceeds 2x10-6
  photons/cm2/sec

28
General Studies

• Localisation
• Stability/accuracy of the likelihood analysis
• Alignment calibration
• Identify warts (data not perfect)
• Studies of background level with orbit position
• Study effect of different event cuts (driven by a
  specific science topic diffuse low background,
  GRB high Aeff etc)
• Dont limit yourself to these suggestions, there
  are many other analysis possibilities.

29
The Next Phase of DC2

• Expect many things to change during DC2 this is
  the beginning, not the end. We expect that
  throughout DC2 there will be
• Updated releases of the software we have
  already identified and fixed a few bugs.
• Updated documentation based on user feedback
• Updated/improved IRFs either because someone
  has come up with a better set of analysis cuts,
  or to improve the implementation into IRFs for
  the cuts we currently use
• Possible reprocessing, if someone comes up with
  an improved analysis that add (or changes)
  variables in the data.

30
Where to from Here?

• At this meeting
• We will describe the details of how the data were
  made and the analyses developed (Wednesday,
  Thursday morning)
• Detailed tutorials on how to perform science
  analyses on the data (Thursday, Friday morning)
• Discuss how the DC2 analysis will be coordinated
  across the large number of DC2 participants
  (Friday).
• DC2 closeout (May 16-18, 2006 at GSFC)
• Describe in detail how the astrophysical sources
  in the sky were defined (this involves detailed
  contributions from many people)
• Contributions and discussions of the analyses
  that were performed during DC2