The SVOM / ECLAIRs Scientific Analysis Pipeline

A. Goldwurm, P. Bacon, N. Bellemont, F. Cangemi, C. Cavet, A. Coleiro, A. Foisseau, A. Gros, C. Lachaud, S. Le Stum, T. Bouchet, J. Rodriguez, L. Bouchet · 2026

Read The SVOM / ECLAIRs Scientific Analysis Pipeline by A. Goldwurm, P. Bacon, N. Bellemont, F. Cangemi, C. Cavet, A. Coleiro, A. Foisseau, A. Gros, C. Lachaud, S. Le Stum, T. Bouchet, J. Rodriguez, L. Bouchet (2026) 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-2604-24254.txt.

Introduction

Goldwurm and colleagues describe the analysis pipeline for ECLAIRs data from the SVOM mission. The paper lays out the overall procedure, the steps, and the main algorithms for analyzing data from the hard X-ray coded mask instrument. The pipeline runs automatically at the science center, generating standard products, and can also run offline with user-chosen parameters. Products and early performance are shown, with ideas for future improvements.

Catalog record

CREATORA. Goldwurm, P. Bacon, N. Bellemont, F. Cangemi, C. Cavet, A. Coleiro, A. Foisseau, A. Gros, C. Lachaud, S. Le Stum, T. Bouchet, J. Rodriguez, L. Bouchet
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE27052
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.24254

Browse the full library · About the Tower of Babel

Full text

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

The SVOM / ECLAIRs Scientific Analysis Pipeline A. Goldwurm, P. Bacon, N. Bellemont, F. Cangemi, C. Cavet, A. Coleiro, A. Foisseau, A. Gros, C. Lachaud, S. Le Stum, T. Bouchet, J. Rodriguez, L. Bouchet Abstract This paper reports on the scientific pipeline for the analysis of the ECLAIRs data of the SVOM mission. We describe the overall procedure, the different steps and the main algorithms of the data analysis for this hard X-ray coded mask instrument. The pipeline runs in automatic mode at the science center generating standard products but can also be used off-line with specific choices of parameters. Generated data products and preliminary performances are illustrated along with perspectives for future improvements. \volnopage Vol.0 (202x) No.0, 000–000 keywords X-rays: general; Techniques: image processing; Software: data analysis. 1 Introduction The ECLAIRs instrument ( Godet et al. 2026 ) on the SVOM space mission ( Cordier et al. 2026 ) is an X-ray / hard X-ray (4-150 keV) telescope based on the coded-mask imaging technique ( Goldwurm & Gros 2022 ) . A position-sensitive detector of 80 \times 80 square CdTe semiconductor pixels records the shadows of a mask projected by the sources in the field of view (FoV). A sky image is reconstructed by correlating the detector image to the mask pattern and by applying cleaning algorithms to reduce the systematic effects. A dedicated ”ECLAIRs pipeline” (ECPI) to perform such an analysis has been developed in the frame of the SVOM French Science Center (FSC) ( Louvin et al. 2026 ) . As the set of X-band telemetry (TM) data, received on ground during a mission download period (pass), are pre-processed at the FSC, ECPI runs automatically in quick look analysis (QLA) mode, feeding the SVOM science database (SDB) with the generated science data products (SDP) to be distributed to the scientists. The pipeline can also be used off-line for a specific analysis on selected data and with user-chosen parameters (energy bands, time intervals, etc.), allowing also the combination of the derived products (stacking, mosaics). The generated SDP are stored in FITS files, compatible with most of the standard astronomical tools (e.g., FTOOLS, DS9, XSPEC) and including all needed metadata (energy bands, exposure times, etc.) in keywords (KW). We describe here the main steps and algorithms of the analysis, mentioning the acronyms of the pipeline components and of the SDP files. The analysis principles are explained in Goldwurm & Gros 2022 and the treatment is similar to the one applied to INTEGRAL / IBIS data ( Goldwurm et al. 2003 ) as the instruments are similar. Notable differences are that the ECLAIRs FoV is much larger ( \approx 90° \times 90°), the Earth often obscures part of it, due to the mission observation strategy, the angular resolution ( \approx 1.5° FWHM) and location accuracy ( < 12 ′ ) are lower, and the background is highly variable, due to the low orbit and frequent passages of the satellite in the energetic particle zones of the South Atlantic Anomaly (SAA). Figure 1: Scheme of the ECLAIRs scientific analysis pipeline. Processing components are shown as rectangles in a sequence from left to right. Input preprocessed TM data are read from the SDB, the auxiliary data (calibration, response and model) from the CalDB. SDP generated by the different components (shown as cylinders in the bottom) are stored in the SDB and may be in input to the following components. The MOSA component is run on SDP produced by the previous components for different passes or observations. 2 Overview of the pipeline The scheme of the pipeline is sketched in Fig. 1 . It has as inputs the SVOM preprocessed TM data stored in the SDB and the set of correction, calibration and response files, prepared by the ECLAIRs Instrument Center and stored in the calibration database (CalDB). ECPI is composed of six large components that run in sequence, with some loops described below. Each component generates SDP, stored in the SDB, which are used by the following components and/or directly by the scientists. The first two of them prepare the data and the correction parameters (DPCO) and calibrate and flag the events (CALI). They can be run only once, and in the most general case, the user can start the analysis from the following component (BUBE), using the SDP stored in the SDB by the automatic processing, since no choice of scientific parameters is made before this level. In the BUBE component (Binning, non-Uniformity, Background, with Earth occultation, correction), based on the input user parameters, the events are selected and binned in detector images, which then are corrected for non-uniformity and background modulation, considering the Earth obscuration. Sky images are then reconstructed and sources are detected, positioned, identified with a catalog source and their ghosts cleaned, in the so called IMAG (imaging) component. To obtain products of specific sources, the BUBE component can be run again to bin detector images properly (small energy bins or time intervals), and then the source product extraction component (SPEX) generates source products, spectra or light curves (LC). The different sets of SDP (for passes or observations) can be combined by stacking detector images and running the analysis components on them, by making sky image mosaics (component MOSA) and combining source spectra, LC and event lists. The ECPI pipeline is implemented in Python and is deployed as a Docker image within the FSC. In online mode, it interacts through an API interface with the other services (databases, pipelines, orchestration). On top of the scientific modules depicted in Fig. 1 , there are other software layers, that deal with data forwarding between scientific components, scientific data product generation and dynamic handling of the ECLAIRs calibration. These layers will be described elsewhere. 3 Data correction and calibration The preprocessed ECLAIRs science data, in input to the data preparation and correction component (DPCO), are in the form of event lists (EVT-CNV files), giving for each recorded single event (multiple are neglected): the time (in UTC), the position Y-Z in the detection plane and the energy channel. Other input TM data are the satellite attitude and orbit, the pixels identified as noisy by the on-board software with the time they were set off/on, and the instrument House-Keeping (HK) data (temperatures, tensions, etc.). ECPI also uses the pre-calculated times of entry and exit of the SAA, both for the core and extended zones. 3.1 Good Time Intervals and Dead/Noisy Pixels DPCO defines the good time intervals (GTI) to be used in the analysis, and establishes the effective live-time of the pixels (PIX-LIF) during the period of the input data. The GTIs are for periods out of SAA passages, of stable attitude (versus unstable or slewing intervals), of no or partial Earth occultation, during no TM gaps and good instrument functioning. Pixel life-time is computed from TM information on the noisy pixels and HK status, and using the CalDB tables of dead pixels (set off permanently). The process also analyzes the science data in order to identify residual dead or noisy pixels, not detected by the on-board software, and to optimize the contours of the SAA based on the recorded count rate. All extracted information is coded in the GTI and PIX-LIF files.

Read the complete text