Neutrino Telescope Development in the West Pacific Ocean

Neutrino Telescope Development in the West Pacific Ocean XXI Workshop on Neutrino Telescopes (NeuTel 2025) Sep.29 - Oct.3, 2025, Padova, Italy Donglian Xu+ Tsung-Dao Lee Institute Oct. 3, 2025 +[email protected] 2 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Astrophysical 𝛎 Sources Neutrino emission from NGC 1068 The high-resolution scan around the most significant location in theNorthernHemisphere is shown in Fig. 2A, with NGC 1068 located inside the 68% confidence region. The position of NGC 1068 produced ^ mns ¼79þ22 #20 more events than expected from the atmospheric and diffuse astrophysicalneutrinobackgrounds. Figure 2B shows the distribution of the angular separation of these events from NGC 1068. Among the 79 most contributing events, 63 were included in a previous analysis (23). The systematic uncertainty on ^ mns is ~2 events (26). The measured spectral index is ^ g¼3:2þ0:2 #0:2 with an estimated systematic uncertainty of ±0.07 (26), consistent with previous results (23). We estimate these systematic uncertainties by analyzing simulated data, assuming a source with flux equal to the one measured for NGC 1068 but varying assumptions about the detector response (26). Systematic uncertainties arise mainly from the modeling of the photon propagation in the glacial ice—e.g., scattering and absorption—and the efficiency with which photons are detected by the IceCube optical modules. Systematic uncertainties are smaller than statistical uncertainties for directional track reconstructions (26)buthaveanonnegligible effect on the energy reconstructions. The properties of the source spectrum are shown in Fig. 3, which shows the likelihood as a function of the model parameters (F 0 ,g) evaluated at the coordinates of NGC 1068. The conversion of ^ mns to the flux F 0 accounts for the contribution from tau neutrino interactions (which produce muons) assuming an equal neutrino flavor ratio. The best-fitting flux averaged over the data-taking period, at a neutrino energy of 1 TeV, is F1Tev nmþ! nm¼ 5:0Tð1:5stat T0:6sysÞ&10#11 TeV#1cm#2s#1. This systematic uncertainty was estimated by varying the flux normalization under different ice and detector properties, such that we reproduce the observed values of ^ gand ^ mns in the median case. Our analysis assumed that the spectrum follows an unbroken powerlawovertheentire energy range of the dataset. However, our results show that the main contribution to the excess (and thus the measured spectral index and flux normalization) comes from neutrinos in an energy range from 1.5 to 15 TeV, which contributes 68% to the total test statistic. Outside this energy range, the data do not strongly constrain the inferred flux properties. Our results strengthen the suggestion (23) that NGC 1068 could be a neutrino source; we find ahigherstatisticalsignificanceforthisresult (4.2sversus 2.9s). Incrementally removing the most contributing neutrino events one by one from the vicinity of NGC 1068 shows that the excess persists, which indicates that it is not dominated by one or a few single events but is the result of an accumulation of neutrinos (26). We visually inspected all neutrino events contributing to the excess from NGC 1068, finding typical, wellreconstructed, horizontal, and approximately tera–electron volt–energy tracks with no sign of unexpected contamination or anomalies (26). Out of the 20 events contributing the most to the test statistic, 19 were included in the previous analysis (23). Although the location is therefore dominated by the same neutrinos, the IceCube Collaboration, Science 378, 538–543 (2022) 4 November 2022 4 of 6 10 −15 10 −12 10−910−610−3100103106 Energy [GeV] 10 −14 10 −13 10 −12 10 −11 10 −10 10 −9 E2Φ[TeV cm−2s−1] IceCube (this work) Theoretical νmodel (52,55) Theoretical νmodel (53) Electromagnetic observations (26) 0.1 to 100 GeV gamma-rays (40,41) > 200 GeV gamma-rays (42) Fig. 4. Multimessenger spectral energy distribution of NGC 1068. Gray points show multifrequency observations (data sources listed in table S1). Dark and light green points indicate gamma-ray observations at 0.1 to 100 GeV (40,41) and >200 GeV (42), respectively. Arrows indicate upper limits, and error bars are 1sconfidence intervals. The solid, dark blue line shows our best-fitting neutrino spectrum with the dark blue shaded region indicating the 95% confidence region. We restrict this spectrum to the range between 1.5 and 15 TeV, where the flux measurement is well constrained (26). Two theoretical predictions are shown for comparison: The light blue shaded region and the gray line show the NGC 1068 neutrino emission models from (52,55) and (53), respectively. The shaded region covers possible values of the gyrofactor 30 ≤h g ≤10 4 used to describe uncertainty in the efficiency of the underlying particle acceleration (55). All fluxes Fare multiplied by the energy squared E 2 . Fig. 5. Comparison of point-source fluxes with the total diffuse astrophysical neutrino flux. Fluxes for NGC 1068 (blue line, this work), TXS 0506+056 (orange line, this work), and the diffuse neutrino background [brown data points and gray band (17,25)] are given for a single flavor of neutrinos and antineutrinos. All fluxes Fvþ! vare multiplied by the neutrino energy squared E2 n. For the conversion of the diffuse astrophysical flux measured from the n e n t channel (17), we assume an equal flavor ratio. Shaded regions and dashed lines indicate 68% confidence intervals. Downward arrows are 68% upper limits. RESEARCH |RESEARCH ARTICLE Downloaded from https://www.science.org at Shanghai Jiao Tong University on November 06, 2022 IceCube Coll., Science 378, 538 (2022) §At least two distinctive categories of sources §Diffuse flux largely unresolved à 1) How to optimize next-gen neutrino telescopes for low and high energies? 2) Room to improve on angular resolution? Need better than 0.1o@ 100TeV to resolve the diffuse flux 3) How to boost flavor identification for discovered sources? 3 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Astrophysical 𝛎 Flavor Identification 8 saturated 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 6): 98.9 pe 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 7): 377.9 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 160 Charge [pe] OM(17, 8): 1253.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 400 Charge [pe] OM(17, 9): 3567.0 pe 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 100 200 300 400 500 Charge [pe] OM(17, 16): 5008.0 pe 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 50 100 150 200 Charge [pe] OM(17, 17): 2372.9 pe 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 Charge [pe] OM(17, 18): 823.0 pe 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 19): 314.7 pe 10.4 10.5 10.6 10.7 10.8 10.9 11.0 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 20): 148.1 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 6): 98.9 pe 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 7): 377.9 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 160 Charge [pe] OM(17, 8): 1253.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 400 Charge [pe] OM(17, 9): 3567.0 pe 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 100 200 300 400 500 Charge [pe] OM(17, 16): 5008.0 pe 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 50 100 150 200 Charge [pe] OM(17, 17): 2372.9 pe 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 Charge [pe] OM(17, 18): 823.0 pe 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 19): 314.7 pe 10.4 10.5 10.6 10.7 10.8 10.9 11.0 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 20): 148.1 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 6): 98.9 pe 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 7): 377.9 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 160 Charge [pe] OM(17, 8): 1253.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 400 Charge [pe] OM(17, 9): 3567.0 pe 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 100 200 300 400 500 Charge [pe] OM(17, 16): 5008.0 pe 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 50 100 150 200 Charge [pe] OM(17, 17): 2372.9 pe 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 Charge [pe] OM(17, 18): 823.0 pe 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 19): 314.7 pe 10.4 10.5 10.6 10.7 10.8 10.9 11.0 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 20): 148.1 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 6): 98.9 pe 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 7): 377.9 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 160 Charge [pe] OM(17, 8): 1253.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 400 Charge [pe] OM(17, 9): 3567.0 pe 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 100 200 300 400 500 Charge [pe] OM(17, 16): 5008.0 pe 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 50 100 150 200 Charge [pe] OM(17, 17): 2372.9 pe 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 Charge [pe] OM(17, 18): 823.0 pe 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 19): 314.7 pe 10.4 10.5 10.6 10.7 10.8 10.9 11.0 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 20): 148.1 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 6): 98.9 pe 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 7): 377.9 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 160 Charge [pe] OM(17, 8): 1253.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 400 Charge [pe] OM(17, 9): 3567.0 pe 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 100 200 300 400 500 Charge [pe] OM(17, 16): 5008.0 pe 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 50 100 150 200 Charge [pe] OM(17, 17): 2372.9 pe 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Time [µs] 0 20 40 60 80 100 120 140 Charge [pe] OM(17, 18): 823.0 pe 10.3 10.4 10.5 10.6 10.7 10.8 10.9 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(17, 19): 314.7 pe 10.4 10.5 10.6 10.7 10.8 10.9 11.0 Time [µs] 0 5 10 15 20 Charge [pe] OM(17, 20): 148.1 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs bright DOM Photoelectrons Time bright DOM °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Exp. Data Fig. 3 Double-cascade event #1 (2012). The reconstructed double-cascade vertex positions are indicated as grey circles, the direction indicated with a grey arrow. The size of the circles illustrates the relative deposited energy, the color encodes relative time (from red to blue). Bright and saturated DOMs are excluded from this analysis. An event view of event #2, observed in 2014 and nicknamed “Double Double,” is shown in Figure 4. The two vertices of the cascades cannot be spatially resolved by eye, highlighting the need for the algorithmic topological classification employed in this work. Analogous to Figure 3,collectedphotoncountsasafunctionof time are displayed together with the predicted photon count distributions for singleand double-cascade hypotheses. The predicted photon count PDFs differ remarkably between the singleand double-cascade hypothesis, with the single-cascade hypothesis disfavored. Data from DOMs labeled as bright were excluded from the analysis, but are used for the comparison of predicted photon count PDFs in Figure 4. Figure 5shows the distribution of the ratio of the double-cascade length Ldc to reconstructed decaycascade energy E2(top panel) and the energy asymmetry AE(bottom panel) of simulated events and data for the best-fit spectrum given in [19]. The distributions were not part of the topological classification chain. While the correlation between Ldc and Etot is clear on average, there are large fluctuations in energy transfer from parent to daughter particle. Therefore, on the per-event basis, the more direct correlation between the double-cascade length Ldc and the decay-cascade energy E2proves more informative. Event #1 has a length-to-energy ratio in a region where the ⌫⌧contribution is larger than the background contribution, but outside of 90% of the simulated ⌫⌧-induced double cascades. Its high energy asymmetry is in a region with 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 2 4 6 8 10 Charge [pe] OM(20, 22): 76.8 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 5 10 15 20 Charge [pe] OM(20, 23): 122.5 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 5 10 15 20 25 30 35 40 Charge [pe] OM(20, 24): 242.0 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 20 40 60 80 100 120 Charge [pe] OM(20, 25): 699.5 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 26): 2175.2 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 27): 1506.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 50 100 150 200 250 Charge [pe] OM(20, 28): 944.1 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 20 40 60 80 100 Charge [pe] OM(20, 29): 579.7 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(20, 30): 337.6 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 2 4 6 8 10 Charge [pe] OM(20, 22): 76.8 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 5 10 15 20 Charge [pe] OM(20, 23): 122.5 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 5 10 15 20 25 30 35 40 Charge [pe] OM(20, 24): 242.0 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 20 40 60 80 100 120 Charge [pe] OM(20, 25): 699.5 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 26): 2175.2 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 27): 1506.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 50 100 150 200 250 Charge [pe] OM(20, 28): 944.1 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 20 40 60 80 100 Charge [pe] OM(20, 29): 579.7 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(20, 30): 337.6 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 2 4 6 8 10 Charge [pe] OM(20, 22): 76.8 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 5 10 15 20 Charge [pe] OM(20, 23): 122.5 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 5 10 15 20 25 30 35 40 Charge [pe] OM(20, 24): 242.0 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 20 40 60 80 100 120 Charge [pe] OM(20, 25): 699.5 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 26): 2175.2 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 27): 1506.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 50 100 150 200 250 Charge [pe] OM(20, 28): 944.1 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 20 40 60 80 100 Charge [pe] OM(20, 29): 579.7 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(20, 30): 337.6 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs Photoelectrons bright DOM bright DOM 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 2 4 6 8 10 Charge [pe] OM(20, 22): 76.8 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 5 10 15 20 Charge [pe] OM(20, 23): 122.5 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 5 10 15 20 25 30 35 40 Charge [pe] OM(20, 24): 242.0 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 20 40 60 80 100 120 Charge [pe] OM(20, 25): 699.5 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 26): 2175.2 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 27): 1506.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 50 100 150 200 250 Charge [pe] OM(20, 28): 944.1 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 20 40 60 80 100 Charge [pe] OM(20, 29): 579.7 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(20, 30): 337.6 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs 10.1 10.2 10.3 10.4 10.5 10.6 10.7 Time [µs] 0 2 4 6 8 10 Charge [pe] OM(20, 22): 76.8 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 5 10 15 20 Charge [pe] OM(20, 23): 122.5 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 5 10 15 20 25 30 35 40 Charge [pe] OM(20, 24): 242.0 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 20 40 60 80 100 120 Charge [pe] OM(20, 25): 699.5 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 26): 2175.2 pe 9.8 9.9 10.0 10.1 10.2 10.3 10.4 Time [µs] 0 50 100 150 200 250 300 350 Charge [pe] OM(20, 27): 1506.9 pe 9.9 10.0 10.1 10.2 10.3 10.4 10.5 Time [µs] 0 50 100 150 200 250 Charge [pe] OM(20, 28): 944.1 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 20 40 60 80 100 Charge [pe] OM(20, 29): 579.7 pe 10.0 10.1 10.2 10.3 10.4 10.5 10.6 Time [µs] 0 10 20 30 40 50 Charge [pe] OM(20, 30): 337.6 pe single cascade double cascade exp. data reco with bright DOMs reco without bright DOMs bright DOM Time °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Data °1.0°0.8°0.6°0.4°0.2 0.0 0.2 0.4 Energy Asymmetry 10°2 10°1 100 Events in 2635 days Single, no brights Double, no brights Double, with brights Exp. Data Fig. 4 Double-cascade event #2 (2014). The reconstructed double-cascade vertex positions are indicated as grey circles, the direction indicated with a grey arrow. The size of the circles illustrates the relative deposited energy, the color encodes relative time (from red to blue). Bright DOMs are excluded from this analysis. Fig. 5 Distribution of the ratio of double-cascade length to reconstructed decay-cascade energy (top), and of the reconstructed energy asymmetry (bottom ) in the double-cascade subsample split by flavor content for the best-fit astrophysical and atmospheric spectra assuming flavor equipartition [19]. The values of the two double cascades are shown. Regions outside of the energy asymmetry values required for double cascades are marked in grey. ray muons, 106νatm, and 104νastro [53–56]. We required that the DOMs on the most illuminated string collected at least 2000 photoelectrons (p. e.) (see Fig. 1) and at least 10 p. e. in the two next-highest-charge, nearest-neighbor strings. Signal events will appear more like cascades than tracks in IceCube, so we also selected events whose morphology was better described by the cascade hypothesis. Aside from 0.6% (22 live days) of the data sample used to confirm agreement between data and simulation (data that were subsequently excluded from our analysis and which contained no signallike events), we performed a “blind”analysis that only used simulated data to devise all selection criteria and analysis methodologies. After application of these initial selection criteria, there was roughly 300 times more background than signal. The expected number of p. e. on the most illuminated string for νastro τCC events after application of these criteria, and additional CNN criteria described below, is shown in Fig. 1. We then created 2D images of DOM number (corresponding to depth) vs time in 3.3 ns bins, with each pixel’s brightness proportional to the digitized waveform amplitude in that time bin. Images were created for the 180 DOMs on the most illuminated string and its two nearest and highestilluminated neighbors, providing three images per event. The image for the highest-charge string on a candidate signal event is shown in Fig. 2(left). The three images were then processed by CNNs, trained to distinguish images produced by simulated signal and background events and based on VGG16 [57], with a total of Oð100 MÞtrainable parameters for the high-dimensional signal parameter space. Three separate CNNs were used to distinguish the ντsignal from remaining backgrounds produced by (i) single cascade neutrino interactions such as νe;μ;τneutral current (NC) and νeCC, (ii) downward-going muons (μ↓), and (iii) both νμ interactions producing muon tracks and μ↓; the associated CNN scores are denoted C1,C2, and C3, respectively, with ranges [0,1]. Figure 2(right) shows SðC1Þ, the saliency [58] for C1, here defined as the magnitude of the gradient of the CNN score (scaled to [0,1]) of C1with respect to the signal amplitude at each pixel. For reference, the contour (solid line) shows where the detected light falls to zero, and is essentially an outline of the plot on the left. (Points outside the contour are variously acausally early, very late, or at distances that are many absorption lengths from the event vertices.) Large SðC1Þvalues indicate where and when changes in light level most effectively change C1. Small SðC1Þvalues appear in highly illuminated regions and in regions with no light. Bright regions contribute to C1, but C1 is not as sensitive there to changes in light level as at the leading and trailing edge envelopes of the light from the event, which are roughly coincident with the contour. The saliency thus shows that C1is sensitive to the overall shape of emitted light in the detector. The scores were calculated for each event, and a signalto-noise ratio of ∼14 was obtained by requiring events to have high scores (C1≥0.99,C2≥0.98 and C3≥0.85). The dominant backgrounds come from other νastro flavors and νatm. The expected energy spectra for signal and the dominant backgrounds, after application of initial and then final selection criteria (including the high CNN scores), are shown in Fig. 3. A subdominant “edge event”background was observed from simulated cosmic-ray muons that deposited most of their Cherenkov light on a single string on the outer edge of the detector. We required C3>0.95 for edge events, reducing this background by about an order of magnitude at an estimated 15% signal loss. Table Ilists the expected FIG. 1. Top: Simulated rate of νastro τCC events binned by the number of p. e. detected by DOMs on the most illuminated string in the event, Qmax str , before any selection criteria (solid) and after the CNN-based criteria (dashed) described in the text. (Downward-going cosmic-ray muons trigger the detector at about 3 kHz, are effectively removed by our selection criteria, and are not shown on the plot; other backgrounds are similarly heavily reduced and also not shown.) Bottom: Ratio of rates (selected or all), showing that signal efficiency grows above about 2000 p.e. The IceCube “GlobalFit”νastro flux [53] is assumed. (Error bars are statistical only.). FIG. 2. Candidate νastro τdetected in September 2015. The left plot shows the DOM number (proportional to depth) versus the time of the digitized PMT signal in 3.3 ns bins for the highestcharge string, with the scale giving the signal amplitude in p.e. in each time bin. The total p.e. detected on the string, Qstr., is shown. The right plot shows SðC1Þ, that string’s saliency map for C1, with darker regions indicating where the C1score is more sensitive to a changing light level (see text). (The Supplemental Material [59] shows three-string views and signed saliencies for all seven νastro τcandidates). PHYSICAL REVIEW LETTERS 132, 151001 (2024) 151001-4 𝝂! IceCube Coll., Eur. Phys. J. C 82, 1031 (2022) M. Meier, J. Soedingrekso, PoS (ICRC2019) 960 L. Wille, D. L. Xu, PoS (ICRC2019) 1036 IceCube Coll., PRL 132, 151001 (2024) w/ machine learning "ν𝒆 IceCube Coll., Nature 591, 220–224 (2021) How can we improve? à 1) Larger detectors for more high-energy events 2) Pixelized DOMs with waveforms for recording kinetic info 4 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) 4 23 KM3NeT KM3NeT: Technical Design Report The cost estimates indicate that for this design a detector containing about 4 units could be constructed. Detectors of such a size will require a certain amount of modularisation. For instance the power required and the data rate produced by such a telescope will require more than one cable running to the shore. Each such cable will require its own primary junction box and seafloor cable network. A detector building block, see Figure 2‐1 that can comfortably be constructed using a single cable network is at most half the size described above. The designs presented and used in the detector performance simulations feature a building block of half the ultimate size. The final performance is scaled accordingly. 2.2 Marine and Earth Sciences The primary objective of the earth and sea science contribution to the KM3NeT programme is to establish a network of detection nodes. This network will incorporate a number of secondary junction boxes strategically positioned around the footprint of the neutrino telescope and connected to a primary junction box. Each secondary junction box will have a suite of sensors connected to it and will deliver continuous realtime data to shore, providing constant long‐time monitoring. Figure 2Ʌ1: Artist’s impression of the neutrino telescope. The number of secondary junction boxes installed will depend on the site, the neutrino telescope footprint and the instrumentation resources required by the science community. The earth and sea science community will use the same shore infrastructure and electrooptical cable for data transfer and power distribution as the neutrino telescope. Medium: Deep-sea water Depth: ~ 3.5 km (ARCA) Volume: ~ 1km! Number of strings : ~230 Medium: Glacial ice Depth: ~ 2.5 km Volume: ~ 8km! Number of strings : ~210 Medium: Deep-sea water Depth: ~ 2.6 km Volume: ~ 1km! Number of strings: ~70 P-ONE (East Pacific Ocean) KM3NeT (Mediterranean Sea) Baikal-GVD (Lake Baikal) Medium: Deep-lake water Depth: ~ 1.4 km Volume: ~ 1km! Number of strings : ~140 IceCube Gen-2 (South Pole) TRIDENT (West Pacific Ocean) Medium: Deep-sea water Depth: ~ 3.5 km Volume: ~ 8km! Number of strings : ~1000 (HUNT, NEON) Next-gen Neutrino Telescopes under Planning Slide edited from W. Tian @ ICRC2025 5 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) TRIDENT Pathfinders : T-REX 2021 September, 2021 https://trident.sjtu.edu.cn/en Pre-selected site conditions §Flat seabed §No nearby high rises or deep trenches §Depth >3km §Close proximity to a shore Hong Kong 540 km 180 km 480 km Sanya Yongxing Island Measured params §Optical properties §Current field §Radioactivity “Hai-Ling Basin” 6 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) T-REX 2021 : Apparatus TRIDENT Preliminary 7 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) T-REX 2021 : Optical Properties §Dedicated analytical and numerical modeling §Exp. data: ~ 1TB Simulated data: ~ 100 TB, 10M files LRM-A LRM-B TRIDENT Coll., Nature Astronomy 7, 1497-1505 (2023) 8 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) T-REX 2021 : Current Field Scaled-down (1:25) experiments in a ship towing tank on SJTU campus Site current field measured on Sep. 6, 2021 Simulation (30-yr): avg. 6 cm/s, max < 26 cm /s 9 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) TRIDENT hybrid DOM : hDOM Array time jitter: < 300ps §Maximize photo-sensitive area & improve timing with SiPMs §Event-by-event tau neutrino identification with PMT waveform readouts Track better than 0.1o @ E𝛎> 100 TeV à TTS: < 2ns 16 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) PMT DOM Camera DOM Light Source Acoustic DOM MuonSLab Real-time cali. cameras: PoS (ICRC2025) 1209 DAQ DOM MuonSLab: PoS (ICRC2025) 440 Acoustic sensor: PoS (ICRC2025) 1102 Buoy PMT DOM 1 PMT DOM 2 Buoy PMT DOM 3 PMT DOM 4 Camera DOM 1 Light Source Camera DOM 2 Acoustic DOM DAQ DOM Buoy Acoustic lock Anchor Oil-filled Optical cable Oil-filled Optical cable Oil-filled Optical cable TRIDENT Pathfinders : T-REX 2024 Deployed in Hai-Ling Basin: 2024.12.01 – 2025.03.26 17 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Preliminary Results from T-REX 2024 Retrieval on March 26th, 2025 Depth~3.3 km, Sea Current < 10 cm/s Real-time calibration cameras : PoS (ICRC2025) 1209 §No visual biofouling for 4-month operation at Hai-Ling Basin §Updated in-situ optical properties measrement with camera system àconsistent with T-REX 2021 results W. Tian et al, NIM-A 1076 (2025) 170489 18 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Preliminary Results from T-REX 2024 TRIDENT Preliminary 40K Ca eTRIDENT Preliminary §Duration: ~O(10 s) §Height: ~10 – 100 × baseline TRIDENT Preliminary Muon Slab J. Wu et al, JINST 20 (2025) P07035 F. Zhang (TAUP 2025, link) §Bursts are mainly casued by marine bioluminescence 𝜟𝒕 dist. for CL2 events Atmo. muon flux @ Hai-Ling Basin 19 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) hDOM Development Progress hDOM motherboard V1: PoS (ICRC2025) 1059 hDOM SiPM: PoS (ICRC2025) 1055 W. Zhi et al, JINST 19 (2024) 06, P06011 ~𝟑𝟎 TTS ~1.8 ns jitter~ 300 ps ~𝟐𝟎 First fully functioning hDOM 3-inch PMT hDOM = Pixelized PMTs (waveforms) + SiPM Arrays (time stamps) FUNDED in 2022 by MOST 20 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) First hDOM-string Deployment Scheduled in Oct. 2025 + TRIDENT Phase-I deployment in 2026 Dedicated calibration string (c-string) under development TRIDENT acoustic positioning system : PoS(ICRC2025) 1102 Ongoing: first hDOM-string assembling & testing 21 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Preliminary Calibration Strategy FUNDED in 2024 by NSFC A standalone multi-purpose calibration device (c-string) for water-based neutrino telescopes §Real-time optical properties monitoring with cameras §Dynamic positioning calibration with acoustic sensors §Bio-luminescence monitoring with wideangle cameras §Angular reconstruction calibration with plastic scintillators 22 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Trigger & DAQ Strategies §High background rate (from K-40) + PMT waveform readout àlarge data throughput •O (10Gbps) for Phase-I §Needs real time data processing and selection at various stages §A combination of hardware (at hDOM motherboard) and software trigger (on-shore cluster) TRIDENT Simulation TRIDENT Simulation TRIDENT trigger strategy: PoS (ICRC2025) 1231 23 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Scientific Program Development Ongoing TRIDENT earth tomography: PoS (ICRC2025) 1059 TRIDENT prompt neutrinos and muons: PoS (ICRC2025) 1062 Also see Iwan’s talk at this conference: link TRIDENT dark matter: X. Xiang (TAUP 2025, link) 24 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Cost-effective Optimization for Scaling up ØOptical Module: hDOM with 3-in vs 4-in PMTs H. Shao et al, A Cost Effective Optimization of the hybrid-DOM Design for TRIDENT [arXiv:2507.10256] ~ 40% reduction on cost & power! 25 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Geometry Optimization for Scaling up I. Morton-Blake et al, Optimizing Underwater Neutrino Telescopes for All-Flavour Point Source Sensitivity (in prep) ØGeomertry layout optimization 1) Sparse vs dense, 2) Clustering vs uniform, 3) Optical properties 32 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) HUNT Two candidate sites: Lake Baikal or South China Sea “Aim for PeVatrons in our Galaxy (>100 TeV)” Slide courtesy: Tianqi Huang 33 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) HUNT : Status & Prospect Activities in South China Sea: Pathfinder: §2 8-in OMs deployed, water depth 1800m (2023.02) Phase I: Long-term monitoring string (2025.01) §4 23-in OMs deployed, water depth 1600m, vertical spacing ~10 m Phase II: Seven-string array (2025-2026) §R&D of OM, APS, waterproof connectors, deployment strategies §Measure the neutrino oscillation parameters HUNT Site Slide courtesy: Tianqi Huang Activities in Lake Baikal: §12 20-in OMs deployed, water depth 1300m (2024.03) §24 23-in OMs deployed, water depth 1300m (2025.03) 34 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) NEON Coverage: 10 km3 Depth: 1700 and 3500 meters 18+1 DOMs per string, with veto Multi (20-31) PMT DOM buoy OM 2 OM 1 anchor 26m 2m Latest trial in Qiandao Lake: Single OM works well under water Slide courtesy: Lili Yang, Sujie Lin 35 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Summary & Outlook §IceCube and KM3NeT have opened a new era for high-energy neutrino astronomy §More telescopes with improved detection ability are needed à resolve origin & increase all-flavor stats for acceleration mechanism studies and BSM searches §New: a viable site was found at a depth of 3.5km in South China Sea for constructing large-scale deep-sea neutrino telescopes àHai-Ling Basin §Multiple independent 𝛎telescope developments are in progress in South China Sea arXiv:2507.10256 §A technology demonstrator (TRIDENT Phase-I) is being developed, anticipated in 2026 §Cost effective optimizations for the TRIDENT big array are under way àTDR Cost effective optimization! Welcome to join us ! 37 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) Optical Modeling of Deep-sea Water Absorption process (𝝀𝒂𝒃𝒔) kill the photons, spacing design Scattering process (𝝀𝒔𝒄𝒂) photon direction, angular resolution Rayleigh scattering (𝜆%&'): Mie scattering (𝜆()*, 𝑐𝑜𝑠𝜃()* ): Attenuation length: 𝐼 𝐿 = 𝐼#) 𝑒$( ' ()*+&' (+,))= 𝐼#) 𝑒$' ()-- F. Hu et. al., Simulation study on the optical processes at deep-sea neutrino telescope sites, NIM-A 1054 (2023) 168367 38 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) T-REX 2021 : Camera system Images captured at depth of 3420m Extended Data Fig. 8: Images of the light emitter at a wavelength of 460 nm with an exposure time of 0.02 s and a water depth of 3420 m. The dashed circle shows the profile of the light emitter and the gray value distribution of the inside region is used for the 2fitting analysis. The region enclosed by the solid circle is used for the analysis of the Icenter method. Left panel: an image taken by CamB at a distance of 21.5 m, the radius of the dashed circle is 65 pixels. Right panel: an image taken by CamA at a distance of 41.6 m, the radius of the dashed circle is 33 pixels. 35 LRM-B: 21.5m LRM-A: 41.6m Camera-calibrating in a ship towing tank W. Tian et. al., A camera system for real-time optical calibration of water-based neutrino telescopes, NIM-A 1076 (2025) 170489 39 Neutrino Telescope Development in the West Pacific Ocean | NeuTel 2025, Sep. 29 – Oct. 3, Padova, Italy Donglian Xu (TDLI) T-REX 2021 : PMT system on the rigid structure, a supporting structure is mounted around the equator of the glass sphere.148 The hydrostatic pressure test is performed for all assembled glass vessel at underwater engineering149 laboratory in Shanghai Jiao Tong University.150 4 PMT Pre-Selection151 Figure 4: The picture of HZC Photonics XP72B22 PMT. XP72B22 PMT produced by Hainan Zhanchuang Photonics Technology Co., Ltd (HZC), shown152 in Figure 4, is chosen for the TRIDENT pathfinder experiment. It is the upgrade of 3-inch XP72B20153 PMT, with optimized shape of glass bulb for an improvement in collection efficiency and transit time154 spread required by JUNO []. The PMT base signal output is connected to the oscilloscope without155 amplification using a coaxial cable and 50 ⌦termination. Several characteristics of XP72B22 PMT156 are listed in Table 2. The typical noise is ⇠1kHz at room temperature with a maximum at 2kHz.157 The typical gain is 3⇥106at a voltage of 1150 V. It has a relatively high quantum efficiency, ⇠20%158 at wavelength of 470 nm.159 Table 2: The description and typical photocathode characteristics of HZC Photonics XP72B22 photomultiplier tube Window material Borosilicate low K Photocathode Bi-alkali Refractive Index at 420nm 1.54 Gain 6⇥103 High Voltage 1150 V Noise 2 kHz, max Quantum Efficiency at 470 nm 20% Quantum Efficiency at 404 nm 28% To select six PMTs, which will be installed in the rDOMs, the individual test and measurement160 for all PMT candidates are carried out at University of Science and Technology of China by () group161 –6– Figure 2:Left: A sketch of receiver module. Optical sensors are housed in lower hemisphere, glued in the optical gel, while electronic boards are installed in upper hemisphere. Right: The 3D-print support structure for PMTs and camera to show the arrangement of optical sensors. system, the dedicated design concept and measurements of the camera system is presented in []109 (cite camera Paper). Three PMTs and one camera are mounted by a 3D-printed support structure110 in the same semisphere, while the necessary electronic boards that manage power supply, analog111 to digital data conversion (ADC) and network nodes for data transmission and slow control are112 integrated in the other semisphere. Moreover, the penetrator and vacuum valve are installed in this113 semisphere. The 3D printed support structure is depicted in the right panel of Figure 2. The three114 3-inch PMTs are arranged with the central axis 30from the vertical line and evenly distributed115 with 120separation. The space of the glass sphere and the PMTs are filled with optical gel, whose116 refractive index is ⇠1.42, to ensure the smooth photon propagation between different media before117 hitting on the PMT photocathod.118 The PMT working voltage was designed at 1375 V, corresponding to a gain at ⇠107. The119 PMT detected signals are firstly pre-amplified and then digitized by an analog-to-digital converter120 (ADC) at a rate of 250 MHz to obtain the waveform. The digitized data is transmitted back to121 the control center on the research vessel through the optical fiber protected by the armoured cable,122 and stored in the real-time processing data acquisition system (DAQ) computer for analyses. Since123 the assessment of the absorption length of the sea water requires a relative measurement between124 two distances, the time synchronization is extremely crucial. Therefore, the clock synchronization125 system between the three optical modules are guaranteed by the White Rabbit technique. The126 design concept of the electronics system is introduced in [] (cite electronics paper).127 3.2 Optical Module Assembly128 The glass vessel is a 17-inch, namely 42 cm diameter VITROVEXrglass sphere. It is composed129 of two independent and identical semispheres, so that the combination of them is flexible. Before130 the enclosure, each semisphere is assembled.131 In one hemisphere, optical sensors are mounted on the 3D-printed structure and then placed132 downwards in the glass. The space between the 3D structure and the glass sphere is filled with133 a highly transparent silicone rubber gel, WACKER SILGELr601 CN, to ensure that the surface134 –4– HZC XP72B22 (3-in) Use relative measurement method to mitigate hidden systematics 21.7 m 41.8 m Figure 1: The detection unit of T-REX. The glass vessel in the middle is equipped with LEDs to mimic an isotropic light source, whereas two glass vessels housing three 3-inch small PMTs and one camera are designed for the light detection. 3 Digital Optical Module92 The digital optical modules is essentially a 17-inch VITROVEXrglass sphere produced by Nautilus93 and equipped with different components. The eDOM has a roughly up-down symmetric structure.94 In each hemisphere, 18 LEDs of three different wavelengths, i.e. 405 (violet), 460 (blue) and 52595 (green) are mounted on the circuit board. Among them, 15 LEDs, namely five for each wavelength96 on the outer edge of the circuit board are designed to be the steady light source and thereby feed97 the camera, whereas three LEDs, namely one for each wavelength in the middle are operated in the98 pulsing mode. A 3D-printed photosensitive resin spherical shell encompasses all the LEDs making99 the emission sphere to be an isotropic light source. The symmetry of two hemispheres of the eDOM100 is moderately violated due to the inevitable mechanical components, penetrator and vacuum valve.101 Therefore, the measurement of LED brightness ratio between two hemispheres is necessary. The102 design concept, testing and calibration of the eDOM are described in [] (cite light source paper).103 3.1 Light Receiver Module104 The two designed rDOM in the pathfinder are essentially identical, except for their orientations.105 The sketch of the rDOM is depicted in the left panel of Figure 2. Each rDOM is equipped with two106 independent light detection system, PMT and camera. These two type of optical sensors correspond107 to the two working modes LEDs, as mentioned above. As this paper will concern with the PMT108 –3– LRM-A LRM-B Electronics: J. N. Tang et. al., Journal of Instrumentation, vol.18 T08001 (2023) M. X. Wang et. al., IEEE Transactions on Nuclear Science, vol. 70, 2240–2247 (2023) Light source: W. L. Li et. al., The Light Source of the TRIDENT Pathfinder Experiment, NIM-A 1056 (2023) 168588