MicroBooNE Detector Research Articles

Convolutional neural network for multiple particle identification in the MicroBooNE liquid argon time projection chamber

We present the multiple particle identification (MPID) network, a convolutional neural network for multiple object classification, developed by MicroBooNE. MPID provides the probabilities that an interaction includes an e−, γ, μ−, π±, and protons in a liquid argon time projection chamber single readout plane. The network extends the single particle identification network previously developed by MicroBooNE [Convolutional neural networks applied to neutrino events in a liquid argon time projection chamber, R. Acciarri et al. J. Instrum. 12, P03011 (2017)]. MPID takes as input an image either cropped around a reconstructed interaction vertex or containing only activity connected to a reconstructed vertex, therefore relieving the tool from inefficiencies in vertex finding and particle clustering. The network serves as an important component in MicroBooNE’s deep-learning-based νe search analysis. In this paper, we present the network’s design, training, and performance on simulation and data from the MicroBooNE detector.29 MoreReceived 23 October 2020Accepted 25 March 2021DOI:https://doi.org/10.1103/PhysRevD.103.092003Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP3.Published by the American Physical SocietyPhysics Subject Headings (PhySH)Physical SystemsNeutrinosTechniquesArtificial neural networksMachine learningParticles & Fields

Vertex-finding and reconstruction of contained two-track neutrino events in the MicroBooNE detector

We describe algorithms developed to isolate and accurately reconstruct two-track events that are contained within the MicroBooNE detector. This method is optimized to reconstruct two tracks of lengths longer than 5cm. This code has applications to searches for neutrino oscillations and measurements of cross sections using quasi-elastic-like charged current events. The algorithms we discuss will be applicable to all detectors running in Fermilab's Short Baseline Neutrino program (SBN), and to any future liquid argon time projection chamber (LArTPC) experiment with beam energies ∼ 1 GeV. The algorithms are publicly available on a GITHUB repository [1]. This reconstruction offers a complementary and independent alternative to the Pandora reconstruction package currently in use in LArTPC experiments, and provides similar reconstruction performance for two-track events.

First Measurement of Differential Charged Current Quasielasticlike ν_{μ}-Argon Scattering Cross Sections with the MicroBooNE Detector.

We report on the first measurement of flux-integrated single differential cross sections for charged-current (CC) muon neutrino (ν_{μ}) scattering on argon with a muon and a proton in the final state, ^{40}Ar (ν_{μ},μp)X. The measurement was carried out using the Booster Neutrino Beam at Fermi National Accelerator Laboratory and the MicroBooNE liquid argon time projection chamber detector with an exposure of 4.59×10^{19} protons on target. Events are selected to enhance the contribution of CC quasielastic (CCQE) interactions. The data are reported in terms of a total cross section as well as single differential cross sections in final state muon and proton kinematics. We measure the integrated per-nucleus CCQE-like cross section (i.e., for interactions leading to a muon, one proton, and no pions above detection threshold) of (4.93±0.76_{stat}±1.29_{sys})×10^{-38} cm^{2}, in good agreement with theoretical calculations. The single differential cross sections are also in overall good agreement with theoretical predictions, except at very forward muon scattering angles that correspond to low-momentum-transfer events.

Scaling the training of particle classification on simulated MicroBooNE events to multiple GPUs

Measurements in Liquid Argon Time Projection Chamber (LArTPC) neutrino detectors, such as the MicroBooNE detector at Fermilab [1], feature large, high fidelity event images. Deep learning techniques have been extremely successful in classification tasks of photographs, but their application to LArTPC event images is challenging, due to the large size of the events. Events in these detectors are typically two orders of magnitude larger than images found in classical challenges, like recognition of handwritten digits contained in the MNIST database or object recognition in the ImageNet database. Ideally, training would occur on many instances of the entire event data, instead of many instances of cropped regions of interest from the event data. However, such efforts lead to extremely long training cycles, which slow down the exploration of new network architectures and hyperparameter scans to improve the classification performance. We present studies of scaling a LArTPC classification problem on multiple architectures, spanning multiple nodes. The studies are carried out on simulated events in the MicroBooNE detector. We emphasize that it is beyond the scope of this study to optimize networks or extract the physics from any results here. Institutional computing at Pacific Northwest National Laboratory and the SummitDev machine at Oak Ridge National Laboratory’s Leadership Computing Facility have been used. To our knowledge, this is the first use of state-of-the-art Convolutional Neural Networks for particle physics and their attendant compute techniques onto the DOE Leadership Class Facilities. We expect benefits to accrue particularly to the Deep Underground Neutrino Experiment (DUNE) LArTPC program, the flagship US High Energy Physics (HEP) program for the coming decades.

Search for heavy neutral leptons decaying into muon-pion pairs in the MicroBooNE detector

We present upper limits on the production of heavy neutral leptons (HNLs) decaying to $\mu \pi$ pairs using data collected with the MicroBooNE liquid-argon time projection chamber (TPC) operating at Fermilab. This search is the first of its kind performed in a liquid-argon TPC. We use data collected in 2017 and 2018 corresponding to an exposure of $2.0 \times 10^{20}$ protons on target from the Fermilab Booster Neutrino Beam, which produces mainly muon neutrinos with an average energy of $\approx 800$ MeV. HNLs with higher mass are expected to have a longer time-of-flight to the liquid-argon TPC than Standard Model neutrinos. The data are therefore recorded with a dedicated trigger configured to detect HNL decays that occur after the neutrino spill reaches the detector. We set upper limits at the $90\%$ confidence level on the element $\lvert U_{\mu4}\rvert^2$ of the extended PMNS mixing matrix in the range $\lvert U_{\mu4}\rvert^2<(6.6$-$0.9)\times 10^{-7}$ for Dirac HNLs and $\lvert U_{\mu4}\rvert^2<(4.7$-$0.7)\times 10^{-7}$ for Majorana HNLs, assuming HNL masses between $260$ and $385$ MeV and $\lvert U_{e 4}\rvert^2 = \lvert U_{\tau 4}\rvert^2 = 0$.

Detector Physics with MicroBooNE

The MicroBooNE detector is a liquid argon time projection chamber (LArTPC), designed for the short-baseline neutrino physics program in the Booster neutrino beamline at Fermilab. Because of their exceptional calorimetric and tracking capabilities, LArTPCs are employed in many current and future neutrino experiments. MicroBooNE, as an operating physics experiment, plays a crucial role in characterising the performance of this technology. We present an overview of the ongoing detector physics studies in MicroBooNE, including a brief introduction to the detector sub-systems and a procedure for calibrating calorimetry in LArTPC. The latter involves studies of signal processing, charge uniformity, ionised electron lifetime and charge recombination. Through the laser system in MicroBooNE, we demonstrate that profound knowledge of the electric field is essential to conduct a neutrino experiment with LArTPCs.

Reconstruction for Liquid Argon TPC Neutrino Detectors Using Parallel Architectures

Neutrinos are particles that interact rarely, so identifying them requires large detectors which produce lots of data. Processing this data with the computing power available is becoming more difficult as the detectors increase in size to reach their physics goals. In liquid argon time projection chambers (TPCs) the charged particles from neutrino interactions produce ionization electrons which drift in an electric field towards a series of collection wires, and the signal on the wires is used to reconstruct the interaction. The MicroBooNE detector currently collecting data at Fermilab has 8000 wires, and planned future experiments like DUNE will have 100 times more, which means that the time required to reconstruct an event will scale accordingly. Modernization of liquid argon TPC reconstruction code, including vectorization, parallelization and code portability to GPUs, will help to mitigate these challenges. The liquid argon TPC hit finding algorithm within the LArSoft framework used across multiple experiments has been vectorized and parallelized. This increases the speed of the algorithm on the order of ten times within a standalone version on Intel architectures. This new version has been incorporated back into LArSoft so that it can be generally used. These methods will also be applied to other low-level reconstruction algorithms of the wire signals such as the deconvolution. The applications and performance of this modernized liquid argon TPC wire reconstruction will be presented.

First Measurement of Inclusive Muon Neutrino Charged Current Differential Cross Sections on Argon at E_{ν}∼0.8 GeV with the MicroBooNE Detector.

We report the first measurement of the double-differential and total muon neutrino charged current inclusive cross sections on argon at a mean neutrino energy of 0.8GeV. Data were collected using the MicroBooNE liquid argon time projection chamber located in the Fermilab Booster neutrino beam and correspond to 1.6×10^{20} protons on target of exposure. The measured differential cross sections are presented as a function of muon momentum, using multiple Coulomb scattering as a momentum measurement technique, and the muon angle with respect to the beam direction. We compare the measured cross sections to multiple neutrino event generators and find better agreement with those containing more complete treatment of quasielastic scattering processes at low Q^{2}. The total flux integrated cross section is measured to be 0.693±0.010(stat)±0.165(syst)×10^{-38} cm^{2}.

Rejecting cosmic background for exclusive charged current quasi elastic neutrino interaction studies with Liquid Argon TPCs; a case study with the MicroBooNE detector

Cosmic ray (CR) interactions can be a challenging source of background for neutrino oscillation and cross-section measurements in surface detectors. We present methods for CR rejection in measurements of charged-current quasielastic-like (CCQE-like) neutrino interactions, with a muon and a proton in the final state, measured using liquid argon time projection chambers (LArTPCs). Using a sample of cosmic data collected with the MicroBooNE detector, mixed with simulated neutrino scattering events, a set of event selection criteria is developed that produces an event sample with minimal contribution from CR background. Depending on the selection criteria used a purity between 50 and 80% can be achieved with a signal selection efficiency between 50 and 25%, with higher purity coming at the expense of lower efficiency. While using a specific dataset and selection criteria values optimized for the MicroBooNE detector, the concepts presented here are generic and can be adapted for various studies of exclusive nu _{mu } CCQE interactions in LArTPCs.

First measurement of νμ charged-current π0 production on argon with the MicroBooNE detector

We report the first measurement of the flux-integrated cross section of $\nu_{\mu}$ charged-current single $\pi^{0}$ production on argon. This measurement is performed with the MicroBooNE detector, an 85 ton active mass liquid argon time projection chamber exposed to the Booster Neutrino Beam at Fermilab. This result on argon is compared to past measurements on lighter nuclei to investigate the scaling assumptions used in models of the production and transport of pions in neutrino-nucleus scattering. The techniques used are an important demonstration of the successful reconstruction and analysis of neutrino interactions producing electromagnetic final states using a liquid argon time projection chamber operating at the earth's surface.

Deep neural network for pixel-level electromagnetic particle identification in the MicroBooNE liquid argon time projection chamber

We have developed a convolutional neural network (CNN) that can make a pixel-level prediction of objects in image data recorded by a liquid argon time projection chamber (LArTPC) for the first time. We describe the network design, training techniques, and software tools developed to train this network. The goal of this work is to develop a complete deep neural network based data reconstruction chain for the MicroBooNE detector. We show the first demonstration of a network's validity on real LArTPC data using MicroBooNE collection plane images. The demonstration is performed for stopping muon and a $\nu_\mu$ charged current neutral pion data samples.

Design and construction of the MicroBooNE Cosmic Ray Tagger system

The MicroBooNE detector utilizes a liquid argon time projection chamber (LArTPC) with an 85 t active mass to study neutrino interactions along the Booster Neutrino Beam (BNB) at Fermilab. With a deployment location near ground level, the detector records many cosmic muon tracks in each beam-related detector trigger that can be misidentified as signals of interest. To reduce these cosmogenic backgrounds, we have designed and constructed a TPC-external Cosmic Ray Tagger (CRT) . This sub-system was developed by the Laboratory for High Energy Physics (LHEP), Albert Einstein center for fundamental physics, University of Bern. The system utilizes plastic scintillation modules to provide precise time and position information for TPC-traversing particles. Successful matching of TPC tracks and CRT data will allow us to reduce cosmogenic background and better characterize the light collection system and LArTPC data using cosmic muons. In this paper we describe the design and installation of the MicroBooNE CRT system and provide an overview of a series of tests done to verify the proper operation of the system and its components during installation, commissioning, and physics data-taking.

Comparison of {varvec{nu }}_{varvec{mu }}-Ar multiplicity distributions observed by MicroBooNE to GENIE model predictions

We measure a large set of observables in inclusive charged current muon neutrino scattering on argon with the MicroBooNE liquid argon time projection chamber operating at Fermilab. We evaluate three neutrino interaction models based on the widely used GENIE event generator using these observables. The measurement uses a data set consisting of neutrino interactions with a final state muon candidate fully contained within the MicroBooNE detector. These data were collected in 2016 with the Fermilab Booster Neutrino Beam, which has an average neutrino energy of 800~hbox {MeV}, using an exposure corresponding to 5.0times 10^{19} protons-on-target. The analysis employs fully automatic event selection and charged particle track reconstruction and uses a data-driven technique to separate neutrino interactions from cosmic ray background events. We find that GENIE models consistently describe the shapes of a large number of kinematic distributions for fixed observed multiplicity.

Scaling studies for deep learning in Liquid Argon Time Projection Chamber event classification

Measurements in Liquid Argon Time Projection Chamber neutrino detectors feature large, high fidelity event images. Deep learning techniques have been extremely successful in classification tasks of photographs, but their application to these event images is challenging, due to the large size of the events, more two orders of magnitude larger than images found in classical challenges like MNIST or ImageNet. This leads to extremely long training cycles, which slow down the exploration of new network architectures and hyperpa-rameter scans to improve the classification performance. We present studies of scaling an LArTPC classification problem on multiple architectures, spanning multiple nodes. The studies are carried out in simulated events in the Micro-BooNE detector.

Ionization electron signal processing in single phase LArTPCs. Part II. Data/simulation comparison and performance in MicroBooNE

The single-phase liquid argon time projection chamber (LArTPC) provides a large amount of detailed information in the form of fine-grained drifted ionization charge from particle traces. To fully utilize this information, the deposited charge must be accurately extracted from the raw digitized waveforms via a robust signal processing chain. Enabled by the ultra-low noise levels associated with cryogenic electronics in the MicroBooNE detector, the precise extraction of ionization charge from the induction wire planes in a single-phase LArTPC is qualitatively demonstrated on MicroBooNE data with event display images, and quantitatively demonstrated via waveform-level and track-level metrics. Improved performance of induction plane calorimetry is demonstrated through the agreement of extracted ionization charge measurements across different wire planes for various event topologies. In addition to the comprehensive waveform-level comparison of data and simulation, a calibration of the cryogenic electronics response is presented and solutions to various MicroBooNE-specific TPC issues are discussed. This work presents an important improvement in LArTPC signal processing, the foundation of reconstruction and therefore physics analyses in MicroBooNE.

Vertex-Finding and Reconstruction for Contained Two-track Neutrino Events in the MicroBooNE Detector

Vertex-Finding and Reconstruction for Contained Two-track Neutrino Events in the MicroBooNE Detector

Simulations for the global quantum efficiency of MicroBooNE optical units

The MicroBooNE detector uses scintillation light from particle interactions in liquid argon as a data acquisition trigger. This scintillation light has wavelengths in the vacuum ultra violet (VUV) range, and must be converted into visible light to be detected by photomultiplier tubes (PMTs). To convert the light, MicroBooNE uses wavelength shifting plates coated with Tetraphenyl butadiene (TPB) placed in front of its PMTs. While basic tuning of this plate-PMT system is sufficient for triggering, precise calibration of the system makes additional calorimetry possible. This note will outline how a photon simulation which accounts for the geometry and optical details of the MicroBooNE detector can accompany a measurement of observed photoelectrons in a plate-PMT test stand, and how the results may be used to determine a “global quantum efficiency” for the plate-PMT system. This global quantum efficiency is one required ingredient for improving the capabilities of the light collection system.

The Pandora multi-algorithm approach to automated pattern recognition of cosmic-ray muon and neutrino events in the MicroBooNE detector

The development and operation of liquid-argon time-projection chambers for neutrino physics has created a need for new approaches to pattern recognition in order to fully exploit the imaging capabilities offered by this technology. Whereas the human brain can excel at identifying features in the recorded events, it is a significant challenge to develop an automated, algorithmic solution. The Pandora Software Development Kit provides functionality to aid the design and implementation of pattern-recognition algorithms. It promotes the use of a multi-algorithm approach to pattern recognition, in which individual algorithms each address a specific task in a particular topology. Many tens of algorithms then carefully build up a picture of the event and, together, provide a robust automated pattern-recognition solution. This paper describes details of the chain of over one hundred Pandora algorithms and tools used to reconstruct cosmic-ray muon and neutrino events in the MicroBooNE detector. Metrics that assess the current pattern-recognition performance are presented for simulated MicroBooNE events, using a selection of final-state event topologies.

Measurement of cosmic-ray reconstruction efficiencies in the MicroBooNE LArTPC using a small external cosmic-ray counter

The MicroBooNE detector is a liquid argon time projection chamber at Fermilab designed to study short-baseline neutrino oscillations and neutrino-argon interaction cross-section. Due to its location near the surface, a good understanding of cosmic muons as a source of backgrounds is of fundamental importance for the experiment. We present a method of using an external 0.5 m (L) × 0.5 m (W) muon counter stack, installed above the main detector, to determine the cosmic-ray reconstruction efficiency in MicroBooNE. Data are acquired with this external muon counter stack placed in three different positions, corresponding to cosmic rays intersecting different parts of the detector. The data reconstruction efficiency of tracks in the detector is found to be ϵdata=(97.1±0.1 (stat) ± 1.4 (sys))%, in good agreement with the Monte Carlo reconstruction efficiency ϵMC = (97.4±0.1)%. This analysis represents a small-scale demonstration of the method that can be used with future data coming from a recently installed cosmic-ray tagger system, which will be able to tag ≈80% of the cosmic rays passing through the MicroBooNE detector.

Michel Electron Reconstruction Using the MicroBooNE LArTPC Cosmic Data

MicroBooNE is a Liquid Argon Time Projection Chamber (LArTPC) neutrino detector located in the Booster Neutrino Beamline at Fermilab which began collecting neutrino data in October 2015. MicroBooNE aims to explore the low energy excess in the νe spectrum reported by MiniBooNE as well as perform ν-Ar cross-section measurements. In this note, we present the current status of reconstructing Michel electrons from cosmic ray muons in the MicroBooNE detector. These electrons are distributed uniformly inside the detector, and serve as a natural and powerful sample to study the detector’s response for low energy (tens of MeV) interactions as a function of position. We have developed a reconstruction software tool to identify such Michel electrons which could be of benefit to LArTPC experiments generically.