Estimating the track-reconstruction efficiency in phenomenological proposals of long-lived-particle searches
Emilie Bertholet, Abner Soffer · 2025
Read Estimating the track-reconstruction efficiency in phenomenological proposals of long-lived-particle searches by Emilie Bertholet, Abner Soffer (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-2501-00857.txt.
Introduction
Emilie Bertholet and Abner Soffer built a software tool for proposed searches for long-lived particles in two thousand twenty-five. Such particles would travel some distance before decaying, leaving displaced tracks in a detector. How well those tracks are rebuilt depends on where the particle started and where it went.
Full detector simulation is heavy and often not open to theory teams. Their package, TrackEff, uses a simplified model of a tracker to count hits for each track. Users then judge if a track would be found. The default setup matches the Belle II drift chamber, but settings can be changed to fit other trackers. The tool helps planners estimate real search reach.
Catalog record
| CREATOR | Emilie Bertholet, Abner Soffer |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 16 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-00857
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Estimating the track-reconstruction efficiency in phenomenological proposals of long-lived-particle searches Emilie Bertholet, Abner Soffer Abstract Phenomenological proposals of searches for new, long-lived particles face a challenge when estimating the reconstruction efficiency of displaced charged-particle tracks. The efficiency depends not only on the charged-particle’s momentum vector, but also on its production point in relation to the detector boundary and substructure elements. Phenomenological studies generally do not have access to GEANT4-based simulation, which is used for calculating the efficiency in experimental analyses. To address this need, we have developed TrackEff, a python software package for estimating the track-reconstruction efficiency. The default detector geometry within TrackEff is that of the Belle II drift chamber. However, tunable parameters enable modification of the geometry to correspond to other tracker configurations. TrackEff uses a simplified model of the tracker to determine the number of detector hits associated with each track. The user can then decide whether the track would be detected based on the number of hits, potentially in association with any other information. 1 Introduction and motivation Recent years have seen a growing interest in searches for new, long-lived particles (LLPs) that arise from physics beyond the standard model, at a variety of experiments. Here we focus on LLPs that decay away from the beam collision point yet inside the tracking volume of the detector. In this case, the tracks of the charged daughters created in the LLP decay emanate from a displaced vertex (DV). The presence of a DV in the signal signature usually provides highly effective background suppression, making such searches experimentally attractive Lee et al. 2019 . When phenomenologists propose a DV-based search, they are faced with the difficult task of estimating the efficiency of the DV reconstruction as a function of its position and other parameters. While the average efficiency for reconstructing promptly produced tracks can be roughly obtained from many published experimental papers, this is generally not the case for a displaced track that originates from a DV. As a result, some phenomenology papers ignore question of efficiency altogether, while others use a simplified parameterization to estimate it. Either approach results in an imprecise estimation of the predicted sensitivity of the proposed search. This calls for a tool for estimating the tracking efficiency that is based on generator-level Monte Carlo without the use of full detector simulation, while accounting for the basic geometric features of the detector. Another use case of such a tool is the early stages of a new detector concept. Construction of a full GEANT4 Agostinelli et al. 2003 model is time consuming. By contrast, a tool that enables easy construction of a simplified detector model can facilitate quick, albeit less accurate, tests of the performance that results from the detector geometry. To address these needs, we have developed the python-based package TrackEff, which is available for use from Zenodo Bertholet and Soffer 2024 . TrackEff can be used for a variety of detector configurations. Nonetheless, since it was developed with the Belle II experiment in mind, we focus on Belle II in what follows. Belle II Abe et al. 2010 ; Altmannshofer et al. 2019 is an asymmetric-energy e^{+}e^{-} B -factory experiment currently operating at SuperKEKB, the worlds highest luminosity collider. With electron and positron beam energies of 7 and 4 GeV, respectively, the experiment typically runs at a center-of-mass energy \sqrt{s}=10.58 GeV, corresponding to the \Upsilon(4S) resonance. Typical studies at Belle II utilize the production of B\bar{B} , \tau^{+}\tau^{-} , or c\bar{c} pairs, as well as other bottomonium or hadronic states. Several DV-based LLP searches have already been conducted at Belle II Adachi et al. 2023 and at its predecessor B factories, the BABAR Lees et al. 2015 and Belle Liventsev et al. 2013 ; Nayak et al. 2024 experiments. Directly related to TrackEff are proposals for LLP searches at such experiments Dey et al. 2021 ; Bertholet et al. 2022 ; Dib et al. 2020 ; Ferber et al. 2022 ; Zhou et al. 2022 ; Kang et al. 2021 ; Duerr et al. 2021 ; Duerr et al. 2020 ; Filimonova et al. 2020 ; Jaeckel and Phan 2023 ; Bandyopadhyay et al. 2022 ; Ferber et al. 2023 . In what follows we describe the algorithm used in TrackEff in Section 2 . Section 3 describes the method by which the algorithm was validated and its parameters were tuned. 2 Estimation of the track reconstruction Efficiency 2.1 Basic idea The reconstruction efficiency is estimated based on the principle that a charged particle will be reconstructed if it left a sufficient number of hits in the tracker. In the case of Belle II, we consider the number of hits in the central drift chamber (CDC) Abe et al. 2010 . Belle and Belle II physics analyses sometimes include an explicit track-quality requirement of 20 CDC hits (e.g., Refs. Adachi et al. 2024 ; Nayak et al. 2024 ). Even without this explicit requirement, the number of hits is used in various stages of the tracking algorithm Bertacchi et al. 2021 . Therefore, to within the approximations of a phenomenological paper, the presence of 20 hits is a reasonable criterion to determine whether a charged-particle track will be reconstructed. TrackEff estimates the number of hits geometrically, as described below. It is left to the user to cut on the desired number of hits. We note that additional considerations may reduce the efficiency to below the TrackEff estimate. In particular, the track-finding algorithm used by the experiment may not be ideal or may have difficulties dealing with large track multiplicities, hard scatters, etc. Additional limitations may exist in vertexing algorithms used to reconstruct DVs from displaced tracks. Given the recent development of an improved tracking algorithm for Belle II Reuter et al. 2024 , these are temporary limitations that can be removed. Since TrackEff is to be used for proposals of new searches and detector designs, it is appropriate to consider its efficiency estimations as what can be achieved with future tracking and vertexing algorithms, even if the algorithms currently in use fall somewhat short of this performance. 2.2 Estimation of the number of hits The default detector-geometry parameters of TrackEff correspond to a simplified model of the Belle II CDC based on Table 6.2 and Fig. 6.2 of Ref. Abe et al. 2010 . The model divides the CDC into 9 radial sections, called superlayers. Each superlayer is divided into a number of layers. The innermost superlayer has 8 layers, and the other remaining superlayers contain 6 layers each. All the layers of a given superlayer are taken to have the same radial width. Each layer may have a different length in the z direction (along the beams). In the default configuration, the lengths are determined from Fig. 6.2 of Ref. Abe et al. 2010 . The actual Belle II CDC has axial and stereo layers, which enables measurement of the z coordinates of hits. However, since this has no bearing on counting the number of hits, in the model all layers are considered axial for simplicity. Each layer contain a specific number of azimuthal cells, varying from 160 for the layers of the innermost superlayer to 384 for outermost superlayer. All the cells