Science

This webpage is kept for archival purposes only and is no longer updated or maintained.

Performance Comparisons Between GLAST Instrument Concepts (Updated DRAFT)

S. W. Digel, Raytheon STX, NASA/GSFC, 11 June 1998

A. Introduction

This document describes simulations that will be used to evaluate the performance of the instrument concepts selected for funding under the GLAST Instrument Technology Development program: the Si strip/CsI calorimeter instrument (PI P. Michelson, Stanford), the scintillating fiber tracker/calorimeter (PI G. Pendleton, UAH), and the Si tracker/converter tower (PI A. Zych, UCR). Their performance relative to the scientific goals of the GLAST mission will be evaluated, with the principal goal being a uniform set of performance evaluations that can be directly compared between instruments. The GLAST Science Requirements document describes basic criteria needed to meet the scientific goals of the mission. The simulations described here will be used to evaluate compliance with these requirements and will also provide performance specifications in terms of the scientific goals. No absolute scoring will be used for the different goals.

The simulations are divided into two categories: instrument simulations, and science simulations. These are similar to the 'Level 1' and 'Level 2' categories of simulations proposed by Hans Mayer-Hasselwander at the Facility Science Team Meeting in September, 1997, without the emphasis on calibration files and scoring. The basic performance of the instruments, i.e., instrumental properties like PSF and energy resolution, will be evaluated using Monte Carlo studies, and the results from these will be used as input to the higher-level simulations of scientific performance. It is important that the basic performance of each instrument be evaluated using the same Monte Carlo simulation package. It is equally important that the basic instrument parameters used for the scientific simulations be based on these results, rather than on scaling arguments that may represent an idealized or only potentially achievable efficiency, angular resolution, etc.

B. Instrument Simulations

Instrument parameters will be derived from event-level simulations with Gismo. The following instrument parameters must be derived and made available as input to the science simulations: (For some scientific goals and for some instruments, it may be necessary to simulate calorimeter-only modes, too. For the Si converter/tracker tower, the parameters below will be evaluated for a single tower, with a specified number of planes, and combined appropriately with the results for the unmodified Si strip/Pb converter towers.)

1. PSF as a function of energy and inclination angle. The parameterization must be detailed enough to include the core and the wings; θ₆₈ and θ ₉₅ are not sufficient. The optimum known reconstruction strategy should be applied, perhaps shared among the instrument designs. The functional form of the PSF (e.g., weighted sum of two Gaussians or sum of Gaussian and exponential) and the grid in energy and inclination angle for which the PSF is to be evaluated will be the same for all instruments. For calorimeter-only modes, the energy range will be restricted appropriately to the higher range of interest for calorimeter-only events. The grid in energy will be 10, 20, 30, 50 MeV then factors of two from 80 MeV to 328 GeV (17 points); in inclination angle the grid will be every 10° from 0 to 80° (9 points).

2. Effective area as a function of energy and inclination angle, for the same grid.

3. Energy resolution as a function of energy and inclination angle, again for the same grid.

4. Background rejection efficiency for charged particles. The rejection efficiency will be judged based on protons and electrons incident on the instruments. The intensities and their angular dependence are TBD, but will be representative of the typical and worst case charged particle backgrounds anticipated for GLAST. If a significant residual background rate or a significant loss of celestial gamma-ray detection efficiency is anticipated after background rejection, the findings will be incorporated in the science simulations as either a background intensity or a reduced effective area.

5. Optimum survey mode/parameters. For each instrument, the optimum survey mode (inertial pointing, zenith pointing, or zenith pointing with rocking) should be identified to maximize exposure or point source sensitivity over the sky (see item 6 below). For a rocking orbit, the parameters of the rocking should be specified. An orbit with 28° inclination and 600 km altitude should be assumed.

C. Science Simulations

The spspecific quantities to be evaluated in the performance simulations are listed below. No ranking of importance is to be inferred from the order of the list. Also, these are not completely independent criteria, in the sense that good performance for one may imply good performance for another, although an attempt has been made to minimize overlap.

6. Point source flux limits and localization for a 2-year sky survey, evaluated for realistic high-latitude and near-plane diffuse backgrounds (to be specified) and a range of source photon spectral indicies from 1.5 to 2.5. Localization regions (95% confidence contour diameter) are to be calculated for a 5-[sigma] source with photon spectral index -2.

7. Limits for flaring sources. The same quantities as above for a 1-orbit time scale, to evaluate the limits for flaring sources. In this case, even with a rocking orbit, the sensitivity will be very non-uniform on the sky, and the flux limits and localization regions should be calculated as a function of inclination angle (0, 20, 40, 60, 80° off-axis).

8. Gamma-ray burst detection & localization - fluence limits and localization as a function of inclination angle for a burst spectrum TBD. If the different instruments have meaningful differences in deadtimes, these should be taken into account here.

9. Sensitivity to spectral breaks - Measurement accuracy and sensitivity to the change in spectral index for a spectral break at specific energies in the 20 MeV-10 GeV range (20, 40, 80, 200, 500 MeV, 1, 2, 4, 8 GeV) for a source of a given flux in a 2-year survey against high-latitude and near-plane backgrounds [TBD].

10. Resolution of extended sources. Minimum angular extent for a source to be resolved (at 99% confidence), for a source of specified flux and spectrum [TBD, representative of a SNR accelerating cosmic rays] observed against a specified background like that near the Galactic plane in the inner Galaxy.

11. Intensity limits for diffuse line emission at 30, 50, and 100 GeV over 3 sr at high latitudes in a 2-year survey. (Calorimeter-only modes may be relevant here.)

D. Additional Comments

Polarization sensitivity may also be evaluated, for source characteristics TBD.

The background rejection efficiency for albedo gamma rays can be defined in terms of the angular resolution far off axis and the parameters of the orbit (including altitude and scanning amplitude) and so albedo rejection is not included as one of the basic simulations.

The basic simulations will be undertaken first, in close coordination with the instrument teams, because all of the science simulations depend on the basic simulations being complete and correct. Evaluating the instrument parameters with the level of detail required is a substantial undertaking. It is hoped that close coordination with the instrument collaborations may result in some optimization of designs before the overall evaluation. That said, in order for the simulations to proceed in a timely fashion, the performance of the readout devices and the trigger and event reconstruction algorithms ideally would be specified and frozen for the simulations. If there is any doubt about the agreement between Gismo and GEANT, then the low-level simulations should be run on both.

The feasibility, schedule, and division of labor for these studies will be discussed at the GLAST Facility Science Team meeting on June 15.

Many important aspects of the instruments are obviously beyond the scope of this study. Mass, power, heat dissipation, noise occupancy, telemetry rate, and cost fall into this category, but are vital to an informed comparison of the instrument concepts.