FermiPhased : A tool for phase-resolved likelihood analysis of Fermi-LAT data

Alexander Lange, B.B. K · 2025

Read FermiPhased : A tool for phase-resolved likelihood analysis of Fermi-LAT data by Alexander Lange, B.B. K (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-04810.txt.

Introduction

Alexander Lange and his co-author present FermiPhased, a tool for phase-resolved study of Fermi telescope data. The Fermi Large Area Telescope tracks high-energy sources across the sky. Pulsars, binaries, and other periodic sources need analysis split by phase.

Built on the Fermipy framework, the tool offers three working modes. Users can bin by phase, hold counts fixed per bin adaptively, or fit several epochs jointly. Parallel execution and cluster support keep large jobs moving. Tutorials and documentation live on GitHub, making phase-resolved fluxes and spectra accessible to more of the gamma-ray community.

Catalog record

CREATORAlexander Lange, B.B. K
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-04810

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

FermiPhased : A tool for phase-resolved likelihood analysis of Fermi-LAT data Alexander Lange, B.B. K Abstract The Fermi Large Area Telescope has enabled detailed studies of high-energy astrophysical sources. To support analysis, we present FermiPhased, a flexible, open-source tool for phase-resolved studies of pulsars, binaries, and other periodically variable sources. Built on the Fermipy framework, FermiPhased offers three modes: standard, adaptive (fixed counts), and joint phase-resolved analysis, enabling users to flexibly bin data based on phase, count statistics, or jointly fit different epochs of interest. FermiPhased is optimized for parallel execution and use on computing clusters. It enables parallelized extraction of phase-resolved fluxes, spectra, and intermediary data products, with tutorials and documentation available on GitHub. I Introduction Since the Large Area Telescope (LAT) launched in 2008 aboard the Fermi Gamma-ray Space Telescope mission, countless variable gamma-ray sources have been discovered. Designed to detect photons between 20 MeV and roughly 1 TeV, Fermi–LAT continuously scans the sky roughly every three hours, providing an all-sky survey of the high-energy universe ( Atwood et al. 2009 ) . The past 17 years have provided a unique opportunity to study known pulsars, and high and low-mass x-ray binaries ( Smith et al. 2023 ) as well as discover previously unknown variable sources Corbet et al. 2019 ; Corbet et al. 2022 and probe the emission mechanisms of high-mass gamma-ray binaries ( Lange et al. 2025a , e.g.,) , gamma-ray pulsars ( Lange et al. 2025b , e.g.,) and microquasars ( Fermi LAT Collaboration et al. 2009 , e.g.,) . FermiTools is the official software developed by the SLAC National Accelerator Laboratory and the Fermi--LAT Collaboration for reducing and analyzing Fermi--LAT data. It includes command-line tools and modules for processing, filtering, and modeling LAT data, supporting tasks such as event selection, and likelihood modeling 1 1 1 For more detailed information and tutorials, please see the FermiTools documentation at: https://fermi.gsfc.nasa.gov/ssc/data/analysis/documentation . . Built on top of FermiTools, Fermipy, is a Python-based high-level analysis package that streamlines much of the tools into easy usage in python. Fermipy ( Wood et al. 2017 ) provides a user-friendly interface for performing source detection, spectral fitting, and generating maps and light curves 2 2 2 For documentation and tutorials, visit: https://fermipy.readthedocs.io/en/latest . . Similarly, there have been user-contributed routines written in various languages ( de Menezes 2022 , e.g., like_bphase.pl 3 3 3 See FermiTools User Contributed Software: https://fermi.gsfc.nasa.gov/ssc/data/analysis/user . and EasyFermi;) to further simplify the reduction and analysis steps 4 4 4 For documentation, visit: https://easyfermi.readthedocs.io/en/latest . . There is, however, a major issue that one can run into, and that is that many analyses can be quite computationally expensive. A phase-resolved (e.g., useful for a spectral changes as a function of pulsar spin or binary orbit) likelihood analysis requires a computationally expensive requirement to generate livetime cubes, exposures maps as the time to complete the analysis is directly related to the number of phase bins. To avoid this, many users have turned to high-performance computing and multi-threading on clustered networks where computational resources are abundant and can easily be split to cater to a high volume of analyses of phase bins (among other things). In this paper, a novel and steamlined procedure to undertake phase-resolved likelihood analysis, FermiPhased , is introduced and its uses explored. In Section II , the basics of the procedure are detailed for the designed GUI. In Section III , case uses of FermiPhased are discussed, including the “standard” test case, a “fixed count” case (useful for pulsar analysis), and a “joint phase-resolved” likelihood fit (useful for sources that exhibit similar behavior like flaring states). In Section IV , the application of FermiPhased for clusters is discussed and in Section V , we discuss the installation and dependencies required to use FermiPhased. In Section VI , we make our concluding remarks. II FermiPhased We introduce the graphical application FermiPhased 5 5 5 More information can be found on Github: https://github.com/aplangrangian/FermiPhased . , released under the Apache License, Version 2.0 (Apache Software Foundation 2004), that is designed to streamline the generation of Fermi–LAT analysis scripts for phase-resolved studies. It offers a user-friendly interface for inputting source parameters—such as right ascension, declination, period, reference epoch (T 0 ), energy range, and time intervals—along with the option to load or save these configurations via JSON files (see Figure 1 ). With labeled input fields and a structured layout, this automated script generator provides shell scripts to call and run FermiTools, create configuration files (. YAML ) for each phase, and a basic FermiPy analysis script. FermiPhased can also be used from the command line, although the graphical interface tool-kit PyQt5 is the focus of this paper. The source code should be slightly modified for the use of remote clusters, for example, using Rivest-Shamir-Adleman (RSA) keys and log-in info with the user’s remote server. The user can browse for spacecraft and event files, specify output directories, and choose to generate a a series of bash scripts ready for execution either on their local machine or a remote cluster. The automatically generated configuration files and analysis scripts to fit the Fermi sources use the user-specified inputs and save outputs such as phase bins, integrated fluxes, spectral parameters, and test statistics. Figure 1: Graphical user interface (GUI) for FermiPhased, a script generator for phase-resolved analysis of Fermi-LAT data. The user can specify key parameters such as source position, energy and time ranges, binning configuration, and file paths. FermiPhased supports multiple modes of phase binning (e.g., Standard Phase-Resolved Analysis, Adaptive (Fixed Counts) Phase-Resolved Analysis, Joint Phase-Resolved Analysis) and allows script export with optional automatic cluster submission for batch processing. Parameters and settings can be saved and loaded here. The scripts call the FermiTools gtmktime , gtselect , gtbin , and gtltcube after accounting for the user inputs. Upon executing these commands, the script performs the likelihood analysis (using FermiPy’s gtlike wrapper) in a generated python file and saves any useful scientific products into a .csv file. For this script to work, the relevant .fits files (photon files and spacecraft files) must be downloaded from the Fermi--LAT data server 6 6 6 More information and caveats at https://fermi.gsfc.nasa.gov/ssc/data/access . Lastly, all analyses are scripted to use the LAT Pass 8 (P8R3_V3) data ( Bruel et al. 2018 ) . Executing the scripts generates a subdirectory, corresponding to each phase bin, within the working directory and starts with gtmktime . The given gtmktime command applies a time filter to Fermi-LAT event data to extract time intervals corresponding to a specific phase bin in a periodic signal (e.g., from a binary system). The filter selects events where both the start and stop times fall within a certain phase window ( \phi ), defined by functions that represent the phase selection. The time is converted from Mission Elapsed Time (MET; in seconds) to Modified Julian Date (MJD), shifted by a reference epoch, T_{0} , and folded on the orbital or rotational period, P (days), to isolate the desired phase slice (of width corresponding to 1/number of bins, n ). For example, for an analysis with 12 phase bins (n=12), the range of phases for the first phase bin is \phi[i=0]=[0.9583,0.04167] , where i=0,...,n-1 .

Read the complete text