[Paper Review] Instrument simulation for the analysis of cosmic ray electron with the Fermi LAT

C. Sgrò, J. Bregeon+1 more|arXiv (Cornell University)|Jul 2, 2009

Dark Matter and Cosmic Phenomena5 references3 citations

TL;DR

This paper presents a detailed Monte Carlo simulation using GEANT4 to model the Fermi LAT instrument response for cosmic ray electron (CRE) analysis, enabling precise event selection and accurate spectrum reconstruction. The simulation achieves sub-10% energy resolution up to 20 GeV and controls systematic uncertainties, resulting in the most precise CRE spectrum measurement from 20 GeV to 1 TeV to date.

ABSTRACT

The Fermi LAT collaboration has built up a detailed Monte Carlo simulation to characterize the instrument response and tune its performance. The simulation code is built around the widely used GEANT4 toolkit and was carefully validated against beam test and flight data. This poster shows how the full LAT simulation is used to develop the event selection for the Cosmic-Ray Electron (\emph{CRE}) analysis so as to optimize the instrument performance. In particular, we will show in detail the determination of the geometry factor and the residual hadron contamination. The very accurate MC simulation proved to be fundamental to control the systematic uncertainties on the CRE spectrum measured by the Fermi LAT.

Motivation & Objective

• To develop a high-fidelity instrument simulation for accurate cosmic ray electron (CRE) spectrum measurement with the Fermi Large Area Telescope (LAT).
• To optimize event selection criteria for CREs by modeling electromagnetic shower topology and distinguishing it from hadronic backgrounds.
• To evaluate and quantify the instrument response functions, including geometry factor and energy resolution, for spectral deconvolution.
• To validate the simulation against beam test and on-orbit data to minimize systematic uncertainties in the CRE spectrum.

Proposed method

• The simulation uses the GEANT4 toolkit to model particle propagation, energy loss, multiple scattering, and energy deposition in all LAT detector components, including active and passive materials.
• A digitization algorithm incorporates realistic detector effects such as gain variations, noise, non-linearities, and sensor misalignments to mirror real data processing.
• The on-orbit particle environment is simulated with a detailed model of cosmic rays, Earth albedo gamma rays, and secondary particles, including geomagnetic cutoff and East-West effects.
• An isotropic electron flux with a power-law energy spectrum (E⁻¹) is simulated to evaluate response functions independently of orbital geometry.
• Event selection is based on a Classification Tree trained on MC data, using shower topology and response features to distinguish electrons from hadrons.
• Instrument response functions—geometry factor and energy resolution—are derived from MC simulations and used to deconvolve the measured event rate into the true CRE spectrum.

Experimental results

Research questions

• RQ1 How accurately can the Fermi LAT simulate the response of cosmic ray electrons across the 20 GeV to 1 TeV energy range using GEANT4?
• RQ2 What are the key detector response features that distinguish electron-induced showers from hadronic showers in the LAT?
• RQ3 How does the simulation-based event selection minimize residual hadron contamination in the CRE sample?
• RQ4 To what extent does the MC simulation control systematic uncertainties in the reconstructed CRE spectrum?
• RQ5 What is the energy resolution and geometry factor of the LAT as a function of true electron energy, as derived from simulation?

Key findings

• The Monte Carlo simulation achieves excellent agreement with beam test data, as shown by the comparison of hit multiplicity distributions in the tracker for 100 GeV/c and 200 GeV/c electron beams.
• The geometry factor increases sharply above 30 GeV due to on-board event selection rejecting events with deposited energy below 20 GeV, which enhances background rejection.
• The energy resolution is better than 10% at 20 GeV and remains below 30% up to 1 TeV, with the 68% containment window centered near unity for reconstructed over true energy.
• The most probable value of the reconstructed energy divided by true energy is consistent with 1 across the entire energy range, confirming an unbiased response up to 1 TeV.
• The residual hadron contamination is minimized through simulation-optimized event selection, validated via comparison with beam and on-orbit data.
• The simulation enables the most precise measurement of the cosmic ray electron spectrum in the 20 GeV to 1 TeV range, with systematic uncertainties controlled by the high-fidelity MC model.