GLAST LAT Performance
Top-level science performance requirements for the LAT are given in Table 1 of the Science Requirements Document
. The resulting performance is governed primarily by three things:
- LAT hardware design
- Event reconstruction algorithms
- Background selections and event quality selections
Thus, although the hardware integration and testing are now complete, as the event selection algorithms are optimized the performance must be updated. In other words, there is not a single, intrinsic set of science performance parameters, but rather results of choices. These are based on detailed Monte Carlo simulations, along with beam tests and ground-level muon tests to check the characterization in the simulation. Standard sets of analysis choices, and their resulting performance characteristics, will be maintained by the LAT team. An earlier performance description is archived here
.
Note that a number of significant improvements are underway, particularly in the background rejection analysis. The performance summarized below is the result of studies for Data Challenge 2 (DC2) in 2006 and will soon be replaced with the results from preparations for flight data analysis.
Instrument Performance
Starting from the front of the instrument, the LAT tracker (TKR) has 12 layers of 3% r.l. tungsten converters (THIN or FRONT section), followed by 4 layers of 18% r.l. tungsten converters (THICK or BACK section). These sections have intrinsically different PSF due to multiple scattering.
A result of the analysis is the production of full instrument response functions (IRFs), describing the performance as a function of photon energy, incidence angle, conversion point within the instrument, and other important parameters. The plots below represent the work of many people on the LAT team. A few important caveats:
- Background rejection analysis is done in bands of energy, resulting in the wiggles. These are under study.
- ClassA events are the result an analysis aimed at the extragalactic diffuse gamma-ray flux measurement, which is the most challenging for background rejection. For that analysis, effective area is sacrificed to obtain the purest sample. For some science topics that do not require that level of background rejection, both classA+classB events may be used. The residual background fraction in the classB events is substantially higher.
- 68% containment of the PSF for class A (black), class A+B (red solid), classA front (red dashed).
|
 |
- Relative effective area vs photon true angle of incidence for 10 GeV photons. The FOV is defined as the integral of effective area over solid angle divided by on-axis effective area. The inflection at 25 degrees is mainly an artifact of the DC2 parameterization.
|
 |
- On-axis effective area, after all selections, for class A+B (solid red line), class A (black line) and ClassA Front (red dashed line) selections. PLEASE NOTE: The post-DC2 analysis, which is now nearing completion, has a smaller reduction in Aeff at high energy. In addition, since the effective area at low energy is particularly influenced by background rejection selections, the LAT team is studying a significantly looser set of selections to increase the effective area below 1 GeV for science topics such as gamma-ray bursts that require far less background rejection level. When that set of selections is standardized, a corresponding set of performance curves will be provided.
|
 |
Point Source Sensitivity Plots
Using the above instrument performance characterization, we have produced additional plots related to point source sensitivity.
The first is a single-energy-bin sensitivity plot, showing the 5-sigma sensitivity to a high-latitude source whose spectrum is integrated over 1/4 decade in energy centered on the energy shown on the horizontal axis. Sensitivity is defined as the flux such that the log of the expected likelihood ratio for detection is 25/2 (or 5 sigma in the Gaussian case) and at least 5 photons. Thus, this plot shows the point source sensitivity using only the photons in each energy bin separately. The assumptions are:
- one calendar year all-sky survey (including effects of the SAA and deadtime)
- diffuse background flux 1.5x10-5 /cm²/s/sr (E>100 MeV), spectral index -2.1
|
|
The point source sensitivity using the information in all energy bins is much better than the individual energy bin sensitivities above. We therefore provide the integral sensitivity measures in two ways.
| First, the bowtie plot, which shows the minimum needed for a 20% determination of the flux after a one-day, one-month, and one-year of operation in all-sky survey for a 1/E² source. The resulting significance at each of these levels is about 8-sigma, the spectral index is determined to about 6%, and the bowtie shape indicates the energy range that contributes the most to the sensitivity. To make a measurement at that level or better, a flat spectral energy density curve must lie above the axis of the bowtie. |
 |
| Finally, experiments are often compared using an integral sensitivity plot (5-sigma sensitivity for E>E0), assuming a 1/E² spectrum source at high latitude. We show here an update for the GLAST LAT: |
 |
If there is any additional information you would find useful to have posted on this page, please contact
S. Ritz
and/or the LAT Analysis Coordinator, Julie Mc Enery.
Useful links:
updated 01 Feb 2007 , maintained by Steve Ritz
