[Paper Review] Baikal-GVD: status and first results

Grigory Safronov|arXiv (Cornell University)|Dec 6, 2020

Astrophysics and Cosmic Phenomena4 citations

TL;DR

Baikal-GVD is a cubic-kilometer-scale deep-underwater neutrino telescope in Lake Baikal, deploying 15 clusters by 2024 to detect high-energy neutrinos from 100 GeV to multi-PeV energies. It reports first results from 2015–2019 data, including 12 downgoing cascade events (>100 TeV) and one upgoing event (~91 TeV), consistent with atmospheric background, and sets a 90% C.L. flux limit on neutrinos from GW170817, validating its sensitivity to astrophysical transients.

ABSTRACT

Baikal-GVD is a cubic-kilometer scale deep-underwater neutrino detector being constructed in Lake Baikal. It is designed to detect neutrinos from $\sim$100 GeV to multi-PeV energies and beyond. Detector deployment began in Spring 2015. Since April 2020 the detector includes seven 8-string clusters carrying in total 2016 optical modules located at depths from 750 to 1275 meters. By the end of the first phase of detector construction in 2024 it is planned to deploy 15 clusters, reaching the effective volume for high-energy cascade detection of 0.75 km$^3$. The design and status of the Baikal-GVD detector and first results of data analysis are presented in this report.

Motivation & Objective

• To construct and deploy a cubic-kilometer-scale neutrino telescope in Lake Baikal to detect high-energy astrophysical neutrinos.
• To validate the detector's performance and background suppression in early operational phases.
• To search for neutrino signals associated with transient astrophysical events such as neutron star mergers.
• To establish the effective detection volume and energy resolution for cascade and muon-like events.
• To contribute to the global search for high-energy cosmic neutrino sources by complementing IceCube and ANTARES.

Proposed method

• The detector uses 15 clusters, each with 8 strings of 36 optical modules (OMs), deployed at depths of 750–1275 m in Lake Baikal.
• Optical modules detect Cherenkov light from charged particles produced in neutrino interactions, with PMTs and digitization at 5 ns resolution.
• Local triggers are generated when adjacent OMs record pulses with charge above thresholds (Q_high: 3–5 p.e., Q_low: 1–2 p.e.) within 100 ns.
• Event data are transmitted to shore via optoelectric cables and processed at JINR for reconstruction using maximum likelihood techniques.
• Background suppression uses 13 track parameters, with Q/ndf as the most powerful discriminant, and zenith angle cuts (θ_zenith > 120°) to select upgoing muons.
• Neutrino energy and direction are reconstructed using time and charge information from pulse shapes, achieving ~4° angular resolution for cascades and ~30% energy uncertainty at 100 TeV.

Experimental results

Research questions

• RQ1 What is the performance of the Baikal-GVD detector in its early deployment phase with seven clusters?
• RQ2 How well does the detector suppress atmospheric muon backgrounds in track-like and cascade event reconstruction?
• RQ3 Are there any significant excesses of high-energy neutrino candidates above atmospheric background in 2015–2019 data?
• RQ4 Can Baikal-GVD detect neutrinos associated with the GW170817 neutron star merger event?
• RQ5 What is the effective volume and sensitivity of Baikal-GVD for high-energy cascade detection?

Key findings

• Fifty-seven track-like neutrino candidate events were selected from April–June 2019, with a background expectation of 54.3 events, showing good agreement with Monte Carlo simulations.
• The median angular resolution for reconstructed track-like events was ~1.0°, indicating high-quality muon reconstruction.
• Twelve downgoing cascade events with reconstructed energies >100 TeV were identified in 2015–2019 data, consistent with atmospheric neutrino expectations.
• One upgoing cascade event with energy ~91 TeV was detected in 2019, suggesting potential astrophysical origin but not yet significant above background.
• No neutrino candidates were found in the direction of NGC 4993 during the GW170817 window, leading to a 90% C.L. flux limit under an E⁻² spectrum and equal flavor flux assumption.
• The effective volume for high-energy cascade detection reached 0.35 km³ in 2020, with a projected 0.75 km³ by the end of Phase 1 in 2024.