gtexpmap Help File

Calculates exposure maps for unbinned likelihood analysis.

Usage: gtexpmap evfile scfile expcube outfile irfs srcrad nlong nlat nenergies

This tool creates exposure maps that are needed to compute the predicted number of photons within a given Region-of-Interest (ROI) for diffuse components in your source model. They are used only for unbinned likelihood analysis. These maps differ significantly from conventional exposure maps, which are the integrals of effective area over time. The exposure calculation used in unbinned likelihood analysis consists of an integral of the total response (effective area times energy dispersion times point spread function) over the entire ROI. See gtlike and gtlike help for more details regarding the likelihood analysis.

The LAT point spread function is relatively broad at low energies. At 100 MeV, 68% of the counts will lie within 3.5 degrees of the source. (See GLAST LAT Performance.) As a consequence, the PSF tails of nearby point sources and diffuse components will overlap significantly with the emission from your sources of interest. To fit your sources accurately, you will also need to simultaneously model the nearby point sources and diffuse components; this will typically require an ROI centered on your sources that is several times the characteristic PSF size in order to have sufficient data to constrain all of the components in your model.

The source model should include any sources that can contribute significantly to the ROI, and again because of the size of the PSF, this implies a "Source Region", centered on the ROI, with a radius that is larger than the ROI radius by several PSF length scales. For example, when fitting a single point source, an ROI with a radius of 10 degrees and a Source Region radius of 20 degrees would be appropriate. Note that since the size of the LAT PSF goes roughly as E^{-0.8}, if
you are considering only higher energy photons (e.g., > 1 GeV), smaller ROI and Source Region radii of just a few degrees may be used.

All of the sources in the Source Region should be included in the source model file that is input to gtlike. (See gtlike, gtlike help, and Model Editor (modeleditor).) The positions and spectral models of these sources can either be obtained from a catalog, such as the Third EGRET catalog, or via a source detection step. The diffuse components will typically include the Galactic diffuse emission and an isotropic extragalactic component, but may also include discrete diffuse components such as SNRs or the LMC.

The exposure map used in the likelihood analysis must extend over the entire Source Region, and be specific to the ROI. The radius of the Source Region is an input to gtexpmap and is given by the srcrad parameter (in degrees). The radius of the ROI is set when the event data are extracted using the gtselect tool, and this information is written to the FITS header of the filtered file. (See gtselect and gtselect Help).

Since we expect that most analyses will consider photon energies down to 100 MeV, we recommend a Source Region radius that is at least 10 degrees larger than the ROI radius, and you will receive a warning message when you run gtexpmap if this is not the case.

The spatial and energy gridding (nlong, nlat, and nenergies parameters) for the Source Region must also be provided in gtexpmap. Half-degree pixels are a nominal choice for gtexpmap, and
that means nlong=120 and nlat=120 if 30 degrees radius was chosen for the srcrad parameter. Smaller pixels should result in a more accurate evaluation of the diffuse source fluxes, but they will also make the exposure map calculation itself more time consuming, scaling roughly
with the number of pixels in your map.

The energy range of the exposure map is determined by the selections made with gtselect to produce the filtered event file, and that energy range is divided into a number of logarithmically spaced bins given by the nenergies parameter. These maps are used to integrate the spectra of the diffuse components in order to determine the predicted counts from these sources. If the spectra of the diffuse components are fairly featureless (i.e., mostly power-laws, with no sharp spectral features such as cut-offs or spectral lines), then 4 or 5 energy bins per decade are probably sufficient.

As input, gtexpmap also needs the livetime spent at each inclination angle at every point in the Source Region. This can be provided by the livetime cube maps, which have the information about the livetime as a function of sky position and off-axis angle. These maps could be created by gtltcube and gtltcube Help or obtained from the SLAC Data Portal. If the livetime cube file is not provided, gtexpmap will calculate these livetimes from the spacecraft file. However, since computing livetimes from the spacecraft file is computationally intensive, it is recommended that you precompute the livetime cubes before running gtexpmap.

Since gtltcube produces a FITS file covering the entire sky, the output of this tool can be used for generating exposure maps for ROIs in other parts of the sky that have the same time interval selections. Note that the livetime cube is calculated on a spatial healpix grid (HEALPix is an acronym for Hierarchical Equal Area isoLatitude Pixelization), while the exposure map is calculated on a longitude-latitude grid.

Examples: gtexpmap

Parameters are passed following the FTOOLs model (i.e., they can be passed interactively by: answering a prompt; as a list in a command line; or by editing the parameter file). To be prompted for gtexpmap parameter values, enter (at the command line): gtexpmap

Note: "Hidden" parameters are not prompted. If you want to overwrite one of the "hidden" parameters, specify the values in the command line. For example, if you do not want to overwrite the output file, set the clobber parameter "clobber=no" ( e.g., gtexmap clobber=no).

An example of how to run gtexpmap is given below:

> gtexpmap The exposure maps generated by this tool are meant
to be used for *unbinned* likelihood analysis only.
Do not use them for binned analyses.
Event data file [] : ps1_55d_filtered.fits
Spacecraft data file [] : spacecraft_data_file.fits
Exposure hypercube file [] : expCube.fits o
utput file name [] : expMap.fits
Response functions [] : P6_V3_DIFFUSE
Radius of the source region (in degrees) [30] : 30
Number of longitude points (2:1000) [120] : 120
Number of latitude points (2:1000) [120] : 120
Number of energies (2:100) [20] : 20
Computing the ExposureMap using expCube.fits

In this case, the source region was selected with gtselect, (see the gtselect help) to be 20 degrees, and the ROI was selected as 30 degrees. The livetime cube was provided (expCube.fits), and the exposure map generated has 120 longitude and latitude points, and 20 energy bins.

The previous example can also be run from the command line as shown below:

>gtexpmap evfile=ps1_55d_filtered.fits scfile=spacecraft_data_file.fits expcube= expCube.fits outfile=expMap.fits irfs=P6_V3_DIFFUSE srcrad=30 nlong=120 nlat=120 nenergies=20


Owned by: J. Chiang

 

Last updated by: Chuck Patterson 04/05/2011