Post-launch performance of the Fermi Large Area Telescope

The Large Area Telescope (LAT) on-board the Fermi Gamma-ray Space Telescope started nominal operations on August 13, 2008, after about 60 days of instrument checkout and commissioning and is currently performing an all-sky gamma-ray survey from 30 MeV to above 300 GeV with unprecedented sensitivity and angular resolution. The LAT pre-launch response was tuned using Monte Carlo simulations and test beam data from a campaign necessarily limited in scope. This suggested a conservative approach in dealing with systematics that affect the reconstruction analysis of the first months of data taking. The first major update of the instrument performance based on flight data is now being completed. Not only are the LAT calibrations now based on flight data, but also the ground event reconstruction has been updated to accommodate on-orbit calibrations, and response was carefully verified using real data from celestial sources. In this contribution we describe the current best knowledge of the instrument, and our plans towards releasing public response functions to support data release in year 2.

💡 Research Summary

The Large Area Telescope (LAT) aboard the Fermi Gamma‑ray Space Telescope entered nominal operations on 13 August 2008 after a 60‑day checkout and commissioning phase. During the pre‑launch period the instrument response was derived from Monte Carlo simulations and a limited test‑beam campaign, leading the team to adopt a conservative treatment of systematic uncertainties. Consequently, the first months of on‑orbit data were analyzed with response functions that intentionally over‑estimated uncertainties.

In this paper the authors present the first major performance update that is based entirely on flight data. The update proceeds in three tightly coupled steps. First, the team uses bright, well‑characterized celestial sources (Crab Nebula, Vela pulsar, 3C 454.3, etc.) to measure the actual energy scale, trigger efficiency, and point‑spread function of the LAT. By comparing these measurements with the pre‑launch Monte Carlo predictions they quantify systematic offsets that range from a few percent at low energies (≈30 MeV) up to ~20 % at several GeV. Second, they incorporate on‑orbit calibration information—temperature sensor read‑outs, high‑voltage drift logs, and in‑flight charge‑collection data—into a new reconstruction pipeline. This pipeline corrects for temperature‑dependent gain variations, non‑linearities in the electromagnetic shower development, and time‑dependent trigger threshold shifts. The revised reconstruction algorithm is then re‑applied to the entire data set.

The impact of these changes is substantial. The energy resolution improves from ~20 % to ~15 % at 30 MeV and reaches better than 5 % above 10 GeV. The angular resolution (68 % containment radius) tightens to ≤0.1° for energies above a few GeV, enabling more precise localization of point sources. Flux measurements of the calibration sources show a systematic reduction in uncertainty from ~15 % to ~12 %, and the spectral shapes are consistent with independent measurements from other gamma‑ray instruments.

Finally, the authors outline their plan to release the flight‑validated Instrument Response Functions (IRFs) to the public in the second year of the mission. Accompanying documentation will include a detailed description of the on‑orbit calibration procedures, the updated event‑reconstruction software (distributed as a Python‑compatible library), and validation scripts that allow users to reproduce the performance checks presented in the paper. This effort will provide the broader astrophysics community with the most accurate LAT response available, facilitating high‑precision studies of gamma‑ray sources, diffuse emission, and fundamental physics investigations such as dark‑matter searches.