高能水下中微子望远镜 (HUNT)

The study of solar neutrinos and supernova neutrinos has greatly advanced the development of astrophysics and particle physics since the 1960s, and nowadays the research on the high-energy neutrinos of astrophysical origin has become one of the frontier hot topics in contemporary astrophysics. High-energy neutrinos are produced in the interaction of high-energy cosmic rays with matter and radiation inside or near their celestial origins. Because neutrinos are electrically neutral, they are not deflected by magnetic fields in space, as cosmic rays are, but can be pointed back directly in the opposite direction to the source from which they were created. As a weakly interacting particle with a very small cross section of interaction with surrounding particles, it can escape from the core of dense celestial objects and propagate freely and unhindered over cosmological distance scales. Therefore, neutrinos are the key messengers in the search for the origin of high-energy cosmic rays. The discovery of cosmic rays began in 1912, and the highest energy of cosmic ray particles observed so far has reached 10²⁰ eV. The energies of these high-energy cosmic rays can exceed the highest particle energies produced by the most powerful particle accelerator on Earth by a factor of ten million. Understanding the physical mechanisms and the extreme physical conditions that make this acceleration process possible is an important step toward understanding the fundamental laws of the universe.

However, the origin of cosmic rays is still unclear and has been a centuries-old problem. The measured cosmic ray energy spectrum has two distinct and important features[1], firstly, the softening of the energy spectrum near 1 PeV, also known as the “knee” structure, and secondly, the hardening of the energy spectrum at 4 EeV, also known as the “ankle” structure. These structures may hint different origins for cosmic rays of different energies. Cosmic rays with energies below the knee region are generally thought to originate from intragalactic objects, while cosmic rays with energies above the ankle region are thought to be dominated by extragalactic objects. The dividing energy between cosmic rays of intragalactic and extragalactic origin, on the other hand, lies between the knee and ankle regions, but the exact value has not yet been determined. Although supernova remnants have been widely accepted as the objects of origin of lower energy cosmic rays with energies in the GeV-TeV range [2], there exists a considerable number of observational phenomena that do not support this class of objects as the origin of higher energy cosmic rays [3]. The search for the origin of PeV cosmic rays in the Milky Way and the origin of higher energy cosmic rays outside the Milky Way, and the understanding of a series of fundamental issues such as their acceleration and propagation are among the most important and hot topics in the field of particle astrophysics and high energy astrophysics at present.

In addition, high-energy neutrinos themselves have the potential to contain significant opportunities for physical results. Their candidate originating objects include black holes, neutron star mergers, stellar explosions, black hole accretion, and other violent astrophysical processes. It is hypothesized that in these phenomena, the gravitational or magnetic energy of the object is converted into the energy of high-energy cosmic rays, which produce high-energy neutrinos in the dense regions of the object. Neutrinos can escape from regions that are opaque to electromagnetic radiation, so they can provide a unique view in the vicinity of exploding stars and black hole horizons, bringing us the information about these most extreme physical processes in the universe. The detection of high-energy neutrinos may also help locate nearby gravitational wave sources and study their astrophysical origins. In addition, achieving a large sample of high-energy neutrino probes themselves means that we can study the nature of weak interactions in detail at high-energy scales, which can further test the Standard Model of particle physics and potentially create opportunities for new physics discoveries.

The U.S. IceCube neutrino telescope at the South Pole detected, for the first time in 2013, neutrino radiation of astrophysical origin with energies above TeV[4]. These high-energy neutrinos are far more energetic than neutrinos from the solar thermonuclear reaction and supernova 1987A neutrinos, ushering in a new era of high-energy neutrino astronomy research. However, the IceCube experiment, through a decade of data accumulation, has only found about a near 4 times significance excess of neutrino events in the direction of two objects, the blazar TXS 0506+06 and the active galaxy NGC 1068[5], respectively. Neither has a clear neutrino source been detected by the same type of experiment internationally, essentially because of the limitations of detector design and cost, which make the effective volume insufficient to the sensitivity is not capable of the detection of the neutrino flow intensity of a single source.

Due to the importance of high-energy neutrino research, several countries have deployed plans to build a new generation of high-energy neutrino detectors. Among them, the most competitive one is the IceCube-Gen2 detector planned to be completed in 2034 by the United States, with a detection volume of 7.9 cubic kilometers. The LHAASO, a national major science and technology infrastructure undertaken by the Institute of High Energy Physics, is the most sensitive device in the ultra-high energy gamma-ray range in the world. Its official scientific operation started in July 2021, and has already achieved important results in the field of ultra-high energy gamma-ray observation, finding PeVatrons all over the Milky Way, providing a number of candidate objects for high-energy neutrino sources. Therefore, the LHAASO team targets high-energy neutrino sources above 100 TeV and takes advantage of the low background of the atmospheric neutrino at this energy range and the lower detector construction cost to propose the construction of an ultra-large scale underwater high-energy neutrino telescope.

The IceCube experiment has so far failed to effectively confirm that high-energy celestial source neutrinos are in agreement with existing observed sources in other wavelengths range, i.e., no high-energy astrophysical neutrino sources have been effectively confirmed. The fundamental reason for this is that the effective detection volume of the experiment is only 1 km³ and the experimental sensitivity is insufficient to detect the neutrino flow intensity of a single source, i.e., it does not have the ability to achieve statistically significant observations of neutrino signals from a single source. the LHAASO experiment currently gives observations of 12 gamma-ray sources above 100 TeV [6], which are potential candidates for the origin of PeV cosmic rays, and observations of neutrinos at 100 TeV and above help to provide decisive evidence for the authentication of the origin of these cosmic rays. With reference to the measurements of the ultra-high energy (>100 TeV) gamma-ray flux of the brightest of these objects, LHAASO J1825-1326, assuming that its gamma-ray flux is entirely produced by hadronic processes, a significant measurement of the source's neutrino radiation within a more feasible observation time (e.g., detection of at least 1 muon neutrino event per year) would require an effective area of muon neutrino detections. This means that the effective detector volume needs to be expanded to ~10³/2 ≈30 times larger than that of the IceCube. Even the largest high-energy neutrino telescope planned to be built, IceCube-Gen2 (about 7.9 times the size of IceCube), would not, under more ideal circumstances, give significant enough observations. Therefore, in order to enhance the effective observation of neutrinos above 100 TeV, we propose a plan to build a high-energy neutrino telescope of up to 30 km³ in Lake Baikal or the South China Sea.

Such a large-scale neutrino telescope would be able to effectively detect not only intragalactic neutrino sources but also extragalactic high-energy neutrino sources, and IceCube has so far found more significant neutrino signals in the direction of the blazar TXS 0506+056 and the active galaxy NGC 1068, giving a confidence level of nearly 3σ. Among them, the neutrino energy spectrum of TXS 0506+056 is relatively hard (the best-fit spectral index is about 2.1), which is more suitable as a target for 100 TeV neutrino observations. And different theoretical models give different physical images for the coincidence event of the blazar TXS 0506+056 with the neutrino event IC-170922A[7]. Under the standard one-zone model of the blazar, the event is a coincidence. If the same event occurs again, the expected number of detection events for a 30 km³ detector is still less than one, i.e., only one neutrino event at most can be detected. However, under the multi-zone model of the blazar, the coincidence can be interpreted as the flare generating a violent energy dissipation in a short time and radiating a neutrino flux sufficient to produce a real coincidence signal. In this case, the IceCube detection is not a small coincidence, and the 30 km³ detector should be able to detect more than one neutrino event if a similar blazar occurs again. Such a detection could effectively distinguish the radiation model of the blazar and help us to determine the correct physical picture and thus understand the acceleration of PeV cosmic rays in active galactic nuclei.

Improving the detection of neutrinos above 100 TeV will also provide more opportunities for multi-messenger observations of cosmic ray sources. IceCube can give real-time warnings of about 12 “golden events” per year, which have a 50% or higher probability of being signals, i.e., neutrinos produced by astrophysical processes, and whose most probable energy is concentrated above 100 TeV. A detector of 30 km³ size is expected to detect about 360 “golden events” per year, among which there should be many events with signal probability much higher than 50%. Multi-messenger observations driven by real-time neutrino alerts can provide additional information for the authentication of neutrino sources, especially for our understanding of PeV cosmic ray acceleration processes in transient sources such as gamma-ray bursts, gravitational wave bursts, nuclear collapse supernovae, and tidal disintegration events.

Under the limitation of cost, the innovative development of core detectors and technologies will be carried out to optimize the design of the high-energy neutrino telescope to achieve the optimal effective detection volume and the most sensitive high-energy neutrino point source detection capability. The detector and technology development work of this project is intended to focus on the following three areas.

The design of the new OM will greatly increase the sensitive optical detection area of a single OM, thus increasing the distance of the OM in the telescope, and ultimately achieving a larger effective detector volume with the same number of OMs. A 20-inch photomultiplier tube (PMT) will be used to replace the 10-inch PMT in the existing IceCube and GVD experiments. If this project is implemented at Lake Baikal, it is expected that the existing 15 m OM spacing can be extended to 36 m and the spacing between strings can reach 130 m for the same signal strength detected. The work in this part includes the development of the 23-inch pressure-resistant glass sphere used to protect the 20-inch PMT, further optimization of the performance of the 20-inch PMT and the LED-based time scale system.

From the results of current research, Lake Baikal has unique natural advantages: pure water quality, flat lake bed, depth of 1300 meters or more, and two months of freezing period per year to start detector installation operations on the ice. The Russian Baikal-GVD group carried out the research work on the underwater neutrino telescope as early as 1990 and accumulated a lot of experience in the field engineering implementation work. According to the Russian side's work experience, the cost of on-site placement is about one tenth of the order of the cost of the string detector itself. Therefore, if this project is implemented at Lake Baikal, the detector construction costs are highly advantageous. At the same time, we need to investigate the feasibility of developing this project in the South China Sea, considering the power supply for the entire detector, the chosen station site needs to be as close to the coast as possible in order to reduce the cost of power cables under the sea. As well as the need for longer-term monitoring of seafloor geology, water quality and currents, etc., to provide a judgment for selecting a suitable station site.

Ray-tracing simulation technology is used to deal with Cherenkov light propagation in particle physics processes. The acceleration is to take advantage of the large number of GPU cores with extremely high independence of light transport tasks to process large volumes of light simultaneously, thus significantly increasing the speed of light transport simulations. This is needed to develop techniques for the simulation of the whole process of neutrino reaction in the medium, the propagation of Cherenkov photons generated by secondary charged particles in water, the light collection of the detector, and the response of the detector, and to complete the optimization, evaluation, and sizing of the detector and structural arrangement. At present, based on the GPU algorithm, the simulation program results for single particle incidence show a 10 times speedup. And the simulation results of GEANT4 have been compared with the real data of GVD.

The following specifications are required for the new generation of neutrino telescopes based on the current observational results and the requirements of multi-messenger studies:

1. The neutrino point source sensitivity is at least 1.5 orders of magnitude higher than that of the current IceCube, achieving an effective observational sensitivity of up to several years for a single source.
   
2. The ability to reconstruct individual high-energy neutrinos in near real-time with sub-degree (<0.50) resolution for follow-up observations and multi-messenger astronomy. The ability to reconstruct individual high-energy neutrinos in the energy range of observations and multi-messenger astronomy, and can be co-analyzed with LHAASO data.
   
3. it has collection rates orders of magnitude higher than the current IceCube array for all types of neutrinos in the energy range 100 TeV to 10 PeV.
   

The basic design scheme of HUNT developed in this project, under the requirements mentioned above, is as follows. In the range of 6 km × 6 km, 2,304 strings are arranged with 24 OMs on each string, the OM spacing within the string is 36 m, and the distance between the strings is 130 m (taking the Baikal station site as an example), i.e., a detector volume of 30 cubic kilometers is achieved. 20-inch PMTs are used for the OMs, with the PMTs facing upward for observation, and the light intensity and time information of the light signal will be measured. Within each OM are integrated LED scaled light sources, high voltage modules, low voltage power supplies, readout electronics, thermohydrometer and other equipment. Between all OMs, the clock is synchronized at 1ns level through the White Rabbit clock distribution system. The readout electronics is stored with 500MHz waveform sampling. A sonar positioning system will be set in the detector to monitor the position of all strings in real time to achieve better than 20cm accuracy. Both LED light source and laser are used to achieve time relative difference scaling for all probes to achieve an accuracy of less than 1ns. The waveform digitization work is realized in OM, and through optical fiber, the data information is transmitted to the upper base station until the shore-based data acquisition system and computer center.

Figure 1 the preliminary Design of the neutrino telescope.

Figure 2 the schematic diagram of DOM.

In order to estimate the feasibility of the experimental scheme, a full simulation software as well as data reconstruction software, containing neutrino interaction simulation, detector detection response simulation, and event direction reconstruction software, were developed in the advance study. The preliminary performance of the experiment is given based on this simulation software.

In our detector scheme, the angular resolution depends mainly on the visible track length for track-type events, including muons produced inside the detector array and muons produced outside the array but entering the array. The angular resolution of 1° can be achieved for muons with a track length of about 900 m, 0.5° for muons with a track length of about 1200 m, 0.1° for muons with a track length of 3000 m, and 0.05° for muons with a track length of 6000 m and crossing the array.

Figure 3 The angular resolution of the three reconstruction methods is shown in the above two plots. The angular resolution of the fast reconstruction is yellow dots, the least-squares method with time residuals is blue dots, and the least-squares method with time residuals with charge weights is red dots. The uplift at 8000 m is caused by the statistical error of the ground sample. In addition, the angular resolution here refers to the median value of the angular deviation of the reconstructed samples.

Figure 4 Estimation of the effective area of the high-energy neutrino telescope to be developed in this project. And the effective area of IceCube [8] is enlarged to 30 times the original size (same zenith angle). The colors represent different zenith angles.

Figure 5 Gamma-ray observations (colored dots) of LHAASO J1825-1326, LHAASO J1849-0003, LHAASO J1956+2845 and HESS J1702-420A and gamma-ray energy spectra (blue and red lines) produced by hadron processes corresponding to different models.

For stable high-energy gamma-ray sources in the Milky Way, it is assumed that the high-energy gamma rays from the LHAASO source are produced by the hadron process of proton-proton reaction, and after the corresponding neutrinos oscillation and the neutrino flavor ratio is close to 1:1:1 at earth. The gamma-ray energy spectra produced by the hadron process are given in Figure 5, and the two colors represent the relatively optimistic and relatively conservative models, corresponding to the neutrino event rate upper and lower limits of the prediction. If the high-energy neutrino telescope to be developed in this project has an effective volume 30 times that of IceCube and is built at Lake Baikal (51.8°N), the event rate of muon neutrinos above 100 TeV in the direction of LHAASO J1825-1326 is expected to be 0.53-1.28 events/year, and that of LHAASO J1849-0003 is about 0.97-2.69 events/year, LHAASO J1956+2845 is about 0.38-0.91 events/year, and HESS J1702-420A is about 3.57-17.5 events/year. If the telescope is built in the South China Sea (16.1°N), the muon neutrino event rates corresponding to the above four sources are 0.35-0.86/year, 0.69-1.89/year, 0.37-0.77/year and 2.62-16.7/year, respectively.

For extragalactic variable sources, the 30 km³ high-energy neutrino telescope is expected to give significant detections and to distinguish different models of flare jet. For example, for the association event of the blazar TXS 0506+056 at the high-energy neutrino IC-170922A, the standard single-zone model, i.e., the single-zone optical meson model [9], the two-zone optical meson model [10], and the two-zone proton-proton collision model [11], predict different fluxes with different neutrino emission durations, as shown in Fig. 6. The standard single-zone model predicts a lower neutrino flux. According to the expectations of this model, the neutrino radiation may be enhanced during the gamma-ray flare of this blazar for a duration of about half a year. According to the effective area simulated in Figure 4, the 30 km³ detector detects only about 1 muon neutrino event in six months. Under the two-zone optical meson model, the neutrino flux can be greatly enhanced and about 14 muon neutrino events can be expected to be detected in half year. The two-zone proton-proton collision model expects a potentially large burst of neutrino radiation over a short period of time, but with a shorter duration of about a few weeks. According to this model, assuming an outburst duration of two weeks, the expected detection of about 17 muon neutrino events is comparable to the two-zone light meson model in total, but the time period of arrival of these events will be very concentrated significantly different from the two-zone light meson model. The difference between these theoretical models lies mainly in the different assumptions about the structure of the jet stream and the surrounding environment, so the 30km³ detector can invert these important physical quantities from neutrino measurements.

The effective area of the detector used for the above event rate predictions is assumed to be proportional to the IceCube by a factor of 30 (see Figure 4). Considering the zenith angle variation due to the Earth's rotation, the effective area for a particular source (fixed declination) is the result of a weighted average of the effective areas for different zenith angles, so the effective area at fixed declination is not strictly equal to 30 times the effective area of IceCube. For example, if the 30 km3 detector is built at Lake Baikal, the effective area for TXS 0506+056 is 14 times larger than that of IceCube at 100 TeV, while for LHAASO J1825-1326 it is more than 100 times larger.

Figure 6 Multiband energy spectrum of the blazar TXS 0506+056 within two weeks of the arrival of neutrino event IC-170922A (yellow data points), and the muon neutrino flux predicted by the different models. The blue solid line is the expectation of the standard single-zone model, i.e., the single-zone optical meson model [9]; the red solid line is the expectation of the two-zone optical meson model [10]; and the green solid line is the expectation of the two-zone proton-proton model with a sustained event of about two weeks [11]. The black solid line is the fitted curve of the literature [10] for multi-band electromagnetic radiation, and the expected radiation energy spectra of the remaining two models in the electromagnetic band are essentially the same as theirs.

Reference：

1. Blandford, R, 2014, Nuclear Physics B, 256, 9
   
2. Ginzburg, V. L. and Syrovatskii, S. I, The Origin of Cosmic Rays, New York: Macmillan, 1964
   
3. Aharonian, F. A., Yang, R.Z. and de Oña Wilhelmi, E., 2019, Nature Astronomy, 3, 561
   
4. IceCube Collaboration, 2013, Science, 342, 1
   
5. IceCube Collaboration, 2020, Phys. Rev. Lett., 124, 051103
   
6. LHAASO Collaboration, 2021, Nature, 594, 33
   
7. IceCube Collaboration, 2018, Science, 361, 146
   
8. IceCube Collaboration, Abbasi, R., Ackermann, M., et al., 2021, arXiv:2101.09836
   
9. Keivani, A., Murase, K., Petropoulou, M., et al., 2018, ApJ, 864, 84
   
10. Xue, R, Liu, R.-Y., Petropoulou, M, et al., 2019, ApJ, 886, 23
    
11. Liu, R.-Y., Wang, K., Xue, R., et al., 2019, Phys. Rev. D, 99, 063008