- Email: [email protected]
High energy neutrino acoustic detection activities in Lake Baikal: Status and results
ARTICLE IN PRESS Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
High energy neutrino acoustic detection activities in Lake Baikal: Status and results V. Aynutdinov a, A. Avrorin a, V. Balkanov a, I. Belolaptikov d, D. Bogorodsky b, N. Budnev b,, I. Danilchenko a, G. Domogatsky a, A. Doroshenko a, A. Dyachok b, Zh.-A. Dzhilkibaev a, S. Fialkovsky f, O. Gaponenko a, K. Golubkov d, O. Gress b, T. Gress b, O. Grishin b, A. Klabukov a, A. Klimov h, A. Kochanov b, K. Konischev d, A. Koshechkin a, V. Kulepov f, L. Kuzmichev c, E. Middell e, S. Mikheyev a, M. Milenin f, R. Mirgazov b, E. Osipova c, G. Pan’kov b, L. Pan’kov b, A. Panfilov a, D. Petukhov a, E. Pliskovsky d, P. Pokhil a, V. Poleschuk a, E. Popova c, V. Prosin c, M. Rozanov g, V. Rubtzov b, A. Sheifler a, A. Shirokov c, B. Shoibonov d, Ch. Spiering e, B. Tarashansky b, R. Wischnewski e, I. Yashin c, V. Zhukov a a
Institute for Nuclear Research, Moscow, Russia Applied Physics Institute of Irkutsk State University, Gagarin blvd. 20, Irkutsk 664003, Russia c Skobeltsyn Institute of Nuclear Physics MSU, Moscow, Russia d Joint Institute for Nuclear Research, Dubna, Russia e DESY, Zeuthen, Germany f Nizhni Novgorod State Technical University, Nizhni Novgorod, Russia g St. Petersburg State Marine University, St. Petersburg, Russia h Kurchatov Institute, Moscow, Russia b
a r t i c l e in f o
a b s t r a c t
Available online 19 March 2009
We review the status of high-energy acoustic neutrino detection activities in Lake Baikal. The Baikal collaboration constructed a hydro-acoustic device which may be regarded as a prototype subunit for a future underwater acoustic neutrino telescope. The device is capable of common operation with the Baikal neutrino telescope NT200+, and is operating at a depth of about 150 m on the ‘‘NT200+ instrumentation string’’. Our measurements show that the integral noise power in the frequency band 20–40 kHz can reach levels as low as about 1 mPa, i.e. one of the lowest noise levels measured at the currently considered acoustic neutrino sites. At the same time, short acoustic pulses with different amplitudes and shapes have been observed. Low sound absorption in Baikal freshwater and absence of strong acoustic noise sources do motivate further activities towards a large-scale acoustic neutrino detector in Lake Baikal. & 2009 Elsevier B.V. All rights reserved.
Keywords: Neutrino telescopes Neutrino astronomy UHE neutrinos Baikal
1. Introduction During the last years, the neutrino astronomy is undergoing rapid development and becomes a new ‘‘window’’ into the Universe. The large scale neutrino telescopes currently under operation (NT200+ [1,2] in Lake Baikal, AMANDA/IceCube [3] at the South Pole and ANTARES [4] in the Mediterranean) detect the Cherenkov light emitted in water or ice by relativistic charged particles produced via neutrino interactions with matter. Back in 1957, Askaryan [5] has shown that a high-energy particle cascade in water, besides the Cherenkov radiation, should also produce an acoustic signal. The potential of the acoustic detection of particle cascades is based on the fact that the
Corresponding author. Tel./fax: +7 395 2 332140.
E-mail address: [email protected] (N. Budnev). 0168-9002/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2009.03.035
absorption length for acoustic waves with a frequency about 30 kHz (the peak frequency of acoustic signals from a shower) in sea water is at least an order of magnitude larger than that of Cherenkov radiation, in the fresh Baikal water this ratio is even close to 100 [6]. The second fact that is favourable for the detection of acoustic signals from cascade showers at distances qof hundreds of metres, or even at such long distances as several kilometres, is that the amplitude of pulses produced by showers in the near-field zone decreases only as the square root of distance, while in the far-field zone it decreases, as the reciprocal of distance from the shower [7,8]. Therefore, in principle, a deep-water acoustic detector of high-energy neutrinos can have a much smaller number of measuring channels compared to a Cherenkov detector with the same effective volume [9]. However, the technology of acoustic detection in high-energy physics is much worse developed than optical methods. Since
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
several years, however, an increasing number of feasibility studies on acoustic particle detection are performed [10–16]. The possibility of practical realization of the idea of acoustic detection depends on both the amplitude of the signal caused by the interaction of a ultrahigh-energy neutrino or other particles with water and the characteristics of acoustic noise in the given water basin. Actually, the possibility of acoustic detection of high-energy neutrinos and the energy detection threshold is determined by the possibility of separating the signals produced by cascade showers from noise produced by other sources.
S131
computer (NOVA-C400 Series) and includes four software programmable amplifiers (PGA-1,2,3,4) and the 16-bit over sampling Analog Devices ADC-AD7722 (ADC-1,2,3,4) with a maximum conversion rate of 0.2 Msamples/s. The single board computer pre-processes the data and communicates with the shore computer via DSL modem and the standard neutrino telescope underwater-shore network [17–19]. The electronics is housed in a cylindrical metallic container with 22 cm outer diameter and 40 cm height. Four hermetic connectors penetrate the upper cap of the container (1-power connection 300 V; 2,3-network twisted-pair cable, 4-coaxial cable for trigger signal from NT200+) and two pairs of hydrophone connectors are arranged on opposite sides of the container.
2. Instrumentation Extraction of small signals from background requires an antenna consisting of a set of hydrophones. The optimum distance between the hydrophones is defined by the condition that it must safely exceed several wavelengths of the expected signal but, on the other hand, should not be too large in order to minimize the number of background impulses captured within the coincidence time window. We have constructed a digital hydro-acoustic device with four input channels, arranged on the corners of a regular tetrahedron, with edge length 1.5 m, as shown in Fig. 1 (left). The module was designed to allow common operation with the Baikal neutrino telescope NT200+ and has been installed in April 2006 at one of the moorings of NT200+. Acoustic signals are recorded by four cylindrical hydrophones H2020C (Hf-1,2,3,4 shown in Fig. 1.) with a sensitivity of about 1 mV/Pa, made from a tangentially polarized piezoceramic CTS-19, and maximum operating depth of about 1000 m. All hydrophones have omni-directional pattern suitable for ambient noise measurements. The signals from the hydrophones are further processed by preamplifiers (Preamp-1,2,3,4) with 70 dB amplification and frequency correction. The acoustic sensor and the sensitivity curves are presented in Figs. 2 and 3. In the range below 1 kHz, the relative amplification is lowered by 20 dB per octave in order to suppress low frequency noise. High frequency noise is suppressed by lowpass switched capacitor filters (LTC1569-6) following the preamplifiers. The further processing is performed by an IC (ADM416x200), which is mounted on the base board (AMBPCI v2.0) of a single board
3. Software and operating modes The module, working under Slackware Linux 11 operating system, provides data acquisition, remote monitoring, and control. There are three operation modes of the instrument: (i) ‘‘Self-trigger’’: Online search for short acoustic pulses of definite shape, which can be interpreted as signals from distant quasi-local sources. (ii) ‘‘External trigger’’: Transmission of a 1 s sample of all hydrophones data to the shore centre, after trigger signal from the neutrino telescope (the central NT200 or the NT200+ outer strings). (iii) An autonomous analysis of acoustic background. The joined operation with NT200+ could give us an opportunity to identify the properties of acoustic emission from cascades and provide energy calibration (assuming that signal strength and flux are high enough and the energy threshold low enough to collect a usable number of true coincidences). To suppress the amount of raw data transferred to the shore station the preprocessing of data is done in situ using the algorithm described in Ref. [20]. The online software is able to detect an acoustic signal which exceeds the threshold and to extract the parameters of the signal, such as maximum amplitude, number of periods, and duration of each period.
Power connection 300 v + network cable Shore station
Underwater part ADM416x200
Twisted Pair
PGA-1
ADC-1
Preamp2
PGA-2
ADC-2
Hf1 Ch.4 Hydrophones
Hf2 Preamp3
Ch.1
PGA-3
ADC-3
Preamp4
.5m
L=1
Hf4
PCI Bus
LAN SBC NOVA-C400
Power Supply
Hf3 Ch.3
Ch.2
Monitor Computer SDSL-16S
AMBPCIv2.0
Wire-cable
Preamp1
PGA-4
ADC-4 -300V
Fig. 1. Schematic view of the underwater 4-channel digital device for detection of acoustic signals from high energy neutrinos.
A Master trigger signal from NT200+ Neutrino Telescope
Antenna A device container
ARTICLE IN PRESS S132
V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
throughout the whole water column. The bottom of a water basin can represent a source of seismic noise and noise associated with gas seeps. Thermal noise represents the lower bound of underwater noise and determines the minimum threshold of acoustic neutrino detection [23]. All the spectra obtained from our measurements can be classed into two groups. The first group includes the spectra with power spectral density (PSD) of noise that decreases more or less uniformly with increasing frequency in the frequency band under study (20–40 kHz). Fig. 4 shows examples of the noise PSD that belong to the first group. This type of noise is produced by multiple sources uniformly distributed near the surface, and is observed under stable and calm meteorological conditions. The qualitative behaviour of noise PSD in Lake Baikal does not contradict the empirical frequency dependence typical for ocean sites [22] with a slope of spectra of about 5–6 dB/octave on average. In winter, the spectra are usually somewhat steeper than in summer. The spectra, which contain some specific features in certain frequency bands, belong to the second group. These spectra are observed under nonstationary, nonuniform meteorological
-100
AS1 AS2 AS3 AS4
Sensitivity, [dB re V/uPa]
-102 -104 -106 -108 -110 -112 -114 -116 -118 -120 15
20
25
30
35
40
Frequency, [KHz] Fig. 3. Sensitivity of acoustic sensors (including 70 dB pre-amplifier gain) in units of dB V/mPa as a function of frequency.
Fig. 2. Acoustic sensor with housed pre-amplifier and hydrophone H2020C.
4. Acoustic noise characteristics at Baikal site Ambient noise in the ocean has been much investigated [21,22]. The most intense sources of noise are waves, wind, rain, ice cracks, ship traffic, etc., which are located in the surface water layer; biological noise and thermal noise are generated
Fig. 4. Examples of the power spectral density of noise (PSD) that belong to the first group (see text).
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
conditions, and are due to ship traffic, gas seeps, and so on [24]. Spectral features may also arise in the presence of some noise sources with specific spectra not far from the measurement site. Fig. 5 shows a data sample of the acoustic noise level in the 20–40 kHz interval, recorded by the 4-channel device. Periods of nonstationary conditions are visible by their high noise contribution. We also note that, in the measurements performed under stable meteorological conditions and in the absence of quasi-local noise sources, we do not observe a relative decrease in the highfrequency spectral components with increasing depth, which is observed in salt water (see, e.g., Ref. [15]). The latter fact is explained by the weaker attenuation of high-frequency waves in fresh water. At any point of time the noise depends on the surface condition rather than on depth. Fig. 6 shows the smoothed time dependences of the acoustic noise amplitude for several depths in the frequency band 22–44 kHz for submersions from the ice cover (several independent measurements in March 2004). Fig. 7 shows RMS of noise distribution. It is seen that at stationary and homogeneous meteorological conditions the
S133
integral noise power in the relevant frequency range 20–40 kHz reaches levels as low as 1 mPa (one of the lowest levels measured at currently considered acoustic neutrino detector sites). The mean value of the acoustic noise is around 2.5 mPa.
5. Pulsed noise characteristics From the assumption that the main mechanism of acoustic signal generation by cascade showers is thermoacoustic [5], it follows that the pressure pulses produced by them should have a bipolar shape. The sign of the first half-wave of a pulse is positive if the water temperature at the site of shower development is higher than the temperature corresponding to maximum water density, TMD, and the heating of water by the cascade leads to its expansion. In the opposite case, the sign of the first half-wave is negative.
Fig. 7. Noise probability (20–40 kHz), as measured with the device in 2006.
5 Fig. 5. Acoustic noise level in the frequency band 20–40 kHz at the depth of 150 m.
4.5
Temperature Profile, 20080321 Temperature of maximal density
4
D = 0050[M] D = 0100[M] D = 0300[M] D = 0500[M] D = 0700[M] D = 0900[M] D = 1100[M]
-30 -35
Temperature, C
RMS of acoustic noise [dB re Pa]
-25
10 mPa
-40 -45
3.5
3
2.5
-50 2
-55 1 mPa
-60
1.5
-65 15
20
25
30
35
40
45
Time [hours]
1
0
200
400
600
800
1000
1200
1400
Depth, m Fig. 6. Time dependence of the acoustic noise in the frequency band 22–44 kHz and depth range 50–1100 m at period of 23–30 March 2004 (local time since midnight of first measurement day).
Fig. 8. Measured temperature as a function of depth (28.3.2008) and temperature of maximal density, TMD (dashed line, calculated from Ref. [26]).
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
The Baikal water temperature at depths below 400 m is very stable around 3:4–3:6 C, see Fig. 8 for a spring 2008 measurement. As also shown in Fig. 8, the temperature is only equal to that of maximal density (TMD—dashed curve calculated using the formula from Ref. [26]) at shallow depths below 200 m; at larger depths they differ significantly, since TMD falls steeply by 0:2 C per 100 m depth. Due to the elongated shape of the showers the acoustic signals are emitted in a pancake-shaped volume with a few degrees opening angle. The expected pulse duration is on the order of several tens of microseconds and is proportional to the shower diameter. The pulse amplitude is proportional to the released energy density. Hence, for determining the properties of acoustic noise as the background against which neutrino induced cascades should be detected, it is necessary to measure the amplitude–time characteristics of this noise in the high-frequency spectral region. Several years of study of acoustic noise in Lake Baikal have shown that the background for acoustic detection of high-energy showers is primarily represented by pulsed noise with short pulse duration [24,25]. The acoustic antenna allows us to estimate the zenith and azimuth angles of incidence of acoustic signals. The position of antenna’s hydrophones in water is fixed by acoustic transponders located around NT200+. Fig. 9 presents an event with bipolar pulses in all four channels, detected by this device. Fig. 10 shows the zenith angle distribution for events with bipolar pulses registered in April–May 2006. Most pulses are located in the vicinity of the horizontal plane (zenith 90 ). Sources of pulses coming from just below horizon are likely located also in the near-surface zone. They appear to come from below horizon due to refraction which is caused by the growth in sound velocity with depth. From the region within 45 around the opposite zenith, i.e. from the deep lake zone below the device, no events with bipolar pulse form have been observed. Fig. 11 presents a comparison between amplitudes for showers of given energy expected at distances of 100 and 1000 m in the Mediterranean Sea and in Lake Baikal, calculated using the formula from Refs. [7,26,27]. Taking into account that the highfrequency noise level in Lake Baikal can be as low as 1 mPa (horizontal line at Fig. 4), it is possible to register showers of energy above 1018 eV from a distance of 100 m, and above 1019 eV
Fig. 10. The distribution of reconstructed zenith angle for events with bipolar pulses.
1 Lake Baikal Mediterranean Sea 100 m 1000 m
10-1 Signal amplitude, Pa
S134
10-2 10-3 10-4 10-5 10-6
1015
1016
1017
1018
1019
1020
1021
E, eV Fig. 11. Dependence of expected acoustic signals amplitude generated by highenergy showers versus shower energy.
from 1000 m. At distances much larger than 1 km, signal amplitudes of showers in Baikal water are similar or higher than in the Mediterranean due to lower sound absorption in Baikal freshwater.
6. Summary
Fig. 9. A triggered event with bipolar pulses (channels 1–4).
A prototype device for detection of acoustic signals from ultrahigh-energy neutrinos in water was constructed and is operating since April 2006 at depth 150 m at the site of the Baikal neutrino telescope. Our measurements show that the integral noise power in the frequency band 20–40 kHz can reach levels as low as about 1 mPa. The main source of the noise including bipolar pulses is the near-surface zone of the lake. Analyzing the sound wave arrival directions, we did not find any bipolar pulses which were generated by sources located at large depth of the lake. Thus, we suggest an acoustic detector design that is based on the deployment of a grid of rather compact acoustic antennas arranged at shallow depths (100–200 m for Lake Baikal), which
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
mainly ‘‘watches’’ the water volume from the surface down to the bottom. We conclude from the ongoing work that favourable conditions for acoustic neutrino detection at Lake Baikal exist clearly motivating further activities.
Acknowledgements This work was supported by the Russian Ministry of Education and Science, the German Ministry of Education and Research, REC-17 BAIKAL and the Russian Fund of Basic Research (Grants 0702-00791, 08-02-00198, 08-02-10001, 08-02-05037, and 08-0205037), and by the Grant of President of Russia NSh-4580.2006.2 and by NATO-Grant NIG-9811707 (2005), and in part by the program ‘‘Development of Scientific Potential in Higher Schools’’ (projects 2.2.1.1/1483, 2.1.1/1539, 2.2.1.1/5901). References [1] V. Aynutdinov, et al., Astropart. Phys. 25 (2006) 140. [2] V. Aynutdinov, et al., Nucl. Instr. and Meth. A 588 (2008) 99 V. Aynutdinov, et al., astro-ph/0811.1109, Nucl. Instr. and Meth. A, in press. [3] A. Achterberg, et al., Astropart. Phys. 26 (2006) 155. [4] J. Carr, ANTARES Collaboration, Nucl. Instr. and Meth. A 588 (2008) 80.
S135
[5] G.A. Askaryan, Atomnaya Energiya 3 (1957) 152. [6] C.S. Clay, H. Medwin, Acoustical Oceanography, Wiley, New York, 1977. [7] G.A. Askaryan, B.A. Dolgoshein, A.N. Kalinovsky, Nucl. Instr. and Meth. A 164 (1979) 267. [8] J.G. Learned, Phys. Rev. D 19 (1979) 3293. [9] G.A. Askaryan, B.A. Dolgoshein, Preprint no. 160, Lebedev Physical Institute of the USSR Academy of Sciences, Moscow (1976); G.A. Askaryan, B.A. Dolgoshein, Pis’ma ZhETF 25 (1977) 232. [10] R. Nahnhauer, S. Bo¨ser (Eds.), in: Proceedings of International Workshop on Acoustic and Radio EeV Neutrino Detection Activities (ARENA2005), World Scientific, Singapore, 2006, p. 298. [11] L.F. Thompson, Nucl. Instr. and Meth. A 588 (2008) 155. [12] S. Danaher, ACoRNE Collaboration, J. Phys. Conf. Ser. 81 (2007) 012011. [13] K. Graf, et al., J. Phys. Conf. Ser. 81 (2007) 012012. [14] G. Riccobene, NEMO Collaboration, J. Phys. Conf. Ser. 81 (2007) 012013. [15] N. Kurahashi, et al., Int. J. Mod. Phys. A 21S1 (2006) 217. [16] S. Boeser, et al., Int. J. Mod. Phys. A 21S1 (2006) 221. [17] V. Aynutdinov, et al., Phys. Atom. Nuclei V 69 (11) (2006) 1914. [18] V. Aynutdinov, et al., Nucl. Instr. and Meth. A 567 (2006) 433. [19] V. Aynutdinov, et al., in: Proceedings of 29th International Cosmic Ray Conference, Pune, India, August 3–10, 2005, astro-ph/0507715. [20] V.M. Aynutdinov, et al., in: Proceedings of the 29th ICRC, vol. 8, Pune, India, 2005, p. 251. [21] V.O. Knudsen, R.S. Alford, J.W. Emling, J. Mar. Res. 7 (1948) 410. [22] A.V. Furduev, in: Ocean Acoustics, Nauka, Moscow, 1974, pp. 615–691 (in Russian). [23] R. Mellen, J. Acoust. Soc. Am. 25 (1952) 478. [24] V. Aynutdinov, et al., Acoust. Phys. 52 (2006) 495. [25] V. Aynutdinov, et al., Int. J. Mod. Phys. A 21S1 (2006) 117. [26] C.T. Chen, F.J. Millero, Limnol. Oceanogr. 31 (1986) 657. [27] R.F. Weiss, E.C. Carmac, V.M. Koropalov, Nature 349 (1991) 665.
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research A journal homepage: www.elsevier.com/locate/nima
High energy neutrino acoustic detection activities in Lake Baikal: Status and results V. Aynutdinov a, A. Avrorin a, V. Balkanov a, I. Belolaptikov d, D. Bogorodsky b, N. Budnev b,, I. Danilchenko a, G. Domogatsky a, A. Doroshenko a, A. Dyachok b, Zh.-A. Dzhilkibaev a, S. Fialkovsky f, O. Gaponenko a, K. Golubkov d, O. Gress b, T. Gress b, O. Grishin b, A. Klabukov a, A. Klimov h, A. Kochanov b, K. Konischev d, A. Koshechkin a, V. Kulepov f, L. Kuzmichev c, E. Middell e, S. Mikheyev a, M. Milenin f, R. Mirgazov b, E. Osipova c, G. Pan’kov b, L. Pan’kov b, A. Panfilov a, D. Petukhov a, E. Pliskovsky d, P. Pokhil a, V. Poleschuk a, E. Popova c, V. Prosin c, M. Rozanov g, V. Rubtzov b, A. Sheifler a, A. Shirokov c, B. Shoibonov d, Ch. Spiering e, B. Tarashansky b, R. Wischnewski e, I. Yashin c, V. Zhukov a a
Institute for Nuclear Research, Moscow, Russia Applied Physics Institute of Irkutsk State University, Gagarin blvd. 20, Irkutsk 664003, Russia c Skobeltsyn Institute of Nuclear Physics MSU, Moscow, Russia d Joint Institute for Nuclear Research, Dubna, Russia e DESY, Zeuthen, Germany f Nizhni Novgorod State Technical University, Nizhni Novgorod, Russia g St. Petersburg State Marine University, St. Petersburg, Russia h Kurchatov Institute, Moscow, Russia b
a r t i c l e in f o
a b s t r a c t
Available online 19 March 2009
We review the status of high-energy acoustic neutrino detection activities in Lake Baikal. The Baikal collaboration constructed a hydro-acoustic device which may be regarded as a prototype subunit for a future underwater acoustic neutrino telescope. The device is capable of common operation with the Baikal neutrino telescope NT200+, and is operating at a depth of about 150 m on the ‘‘NT200+ instrumentation string’’. Our measurements show that the integral noise power in the frequency band 20–40 kHz can reach levels as low as about 1 mPa, i.e. one of the lowest noise levels measured at the currently considered acoustic neutrino sites. At the same time, short acoustic pulses with different amplitudes and shapes have been observed. Low sound absorption in Baikal freshwater and absence of strong acoustic noise sources do motivate further activities towards a large-scale acoustic neutrino detector in Lake Baikal. & 2009 Elsevier B.V. All rights reserved.
Keywords: Neutrino telescopes Neutrino astronomy UHE neutrinos Baikal
1. Introduction During the last years, the neutrino astronomy is undergoing rapid development and becomes a new ‘‘window’’ into the Universe. The large scale neutrino telescopes currently under operation (NT200+ [1,2] in Lake Baikal, AMANDA/IceCube [3] at the South Pole and ANTARES [4] in the Mediterranean) detect the Cherenkov light emitted in water or ice by relativistic charged particles produced via neutrino interactions with matter. Back in 1957, Askaryan [5] has shown that a high-energy particle cascade in water, besides the Cherenkov radiation, should also produce an acoustic signal. The potential of the acoustic detection of particle cascades is based on the fact that the
Corresponding author. Tel./fax: +7 395 2 332140.
E-mail address: [email protected] (N. Budnev). 0168-9002/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.nima.2009.03.035
absorption length for acoustic waves with a frequency about 30 kHz (the peak frequency of acoustic signals from a shower) in sea water is at least an order of magnitude larger than that of Cherenkov radiation, in the fresh Baikal water this ratio is even close to 100 [6]. The second fact that is favourable for the detection of acoustic signals from cascade showers at distances qof hundreds of metres, or even at such long distances as several kilometres, is that the amplitude of pulses produced by showers in the near-field zone decreases only as the square root of distance, while in the far-field zone it decreases, as the reciprocal of distance from the shower [7,8]. Therefore, in principle, a deep-water acoustic detector of high-energy neutrinos can have a much smaller number of measuring channels compared to a Cherenkov detector with the same effective volume [9]. However, the technology of acoustic detection in high-energy physics is much worse developed than optical methods. Since
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
several years, however, an increasing number of feasibility studies on acoustic particle detection are performed [10–16]. The possibility of practical realization of the idea of acoustic detection depends on both the amplitude of the signal caused by the interaction of a ultrahigh-energy neutrino or other particles with water and the characteristics of acoustic noise in the given water basin. Actually, the possibility of acoustic detection of high-energy neutrinos and the energy detection threshold is determined by the possibility of separating the signals produced by cascade showers from noise produced by other sources.
S131
computer (NOVA-C400 Series) and includes four software programmable amplifiers (PGA-1,2,3,4) and the 16-bit over sampling Analog Devices ADC-AD7722 (ADC-1,2,3,4) with a maximum conversion rate of 0.2 Msamples/s. The single board computer pre-processes the data and communicates with the shore computer via DSL modem and the standard neutrino telescope underwater-shore network [17–19]. The electronics is housed in a cylindrical metallic container with 22 cm outer diameter and 40 cm height. Four hermetic connectors penetrate the upper cap of the container (1-power connection 300 V; 2,3-network twisted-pair cable, 4-coaxial cable for trigger signal from NT200+) and two pairs of hydrophone connectors are arranged on opposite sides of the container.
2. Instrumentation Extraction of small signals from background requires an antenna consisting of a set of hydrophones. The optimum distance between the hydrophones is defined by the condition that it must safely exceed several wavelengths of the expected signal but, on the other hand, should not be too large in order to minimize the number of background impulses captured within the coincidence time window. We have constructed a digital hydro-acoustic device with four input channels, arranged on the corners of a regular tetrahedron, with edge length 1.5 m, as shown in Fig. 1 (left). The module was designed to allow common operation with the Baikal neutrino telescope NT200+ and has been installed in April 2006 at one of the moorings of NT200+. Acoustic signals are recorded by four cylindrical hydrophones H2020C (Hf-1,2,3,4 shown in Fig. 1.) with a sensitivity of about 1 mV/Pa, made from a tangentially polarized piezoceramic CTS-19, and maximum operating depth of about 1000 m. All hydrophones have omni-directional pattern suitable for ambient noise measurements. The signals from the hydrophones are further processed by preamplifiers (Preamp-1,2,3,4) with 70 dB amplification and frequency correction. The acoustic sensor and the sensitivity curves are presented in Figs. 2 and 3. In the range below 1 kHz, the relative amplification is lowered by 20 dB per octave in order to suppress low frequency noise. High frequency noise is suppressed by lowpass switched capacitor filters (LTC1569-6) following the preamplifiers. The further processing is performed by an IC (ADM416x200), which is mounted on the base board (AMBPCI v2.0) of a single board
3. Software and operating modes The module, working under Slackware Linux 11 operating system, provides data acquisition, remote monitoring, and control. There are three operation modes of the instrument: (i) ‘‘Self-trigger’’: Online search for short acoustic pulses of definite shape, which can be interpreted as signals from distant quasi-local sources. (ii) ‘‘External trigger’’: Transmission of a 1 s sample of all hydrophones data to the shore centre, after trigger signal from the neutrino telescope (the central NT200 or the NT200+ outer strings). (iii) An autonomous analysis of acoustic background. The joined operation with NT200+ could give us an opportunity to identify the properties of acoustic emission from cascades and provide energy calibration (assuming that signal strength and flux are high enough and the energy threshold low enough to collect a usable number of true coincidences). To suppress the amount of raw data transferred to the shore station the preprocessing of data is done in situ using the algorithm described in Ref. [20]. The online software is able to detect an acoustic signal which exceeds the threshold and to extract the parameters of the signal, such as maximum amplitude, number of periods, and duration of each period.
Power connection 300 v + network cable Shore station
Underwater part ADM416x200
Twisted Pair
PGA-1
ADC-1
Preamp2
PGA-2
ADC-2
Hf1 Ch.4 Hydrophones
Hf2 Preamp3
Ch.1
PGA-3
ADC-3
Preamp4
.5m
L=1
Hf4
PCI Bus
LAN SBC NOVA-C400
Power Supply
Hf3 Ch.3
Ch.2
Monitor Computer SDSL-16S
AMBPCIv2.0
Wire-cable
Preamp1
PGA-4
ADC-4 -300V
Fig. 1. Schematic view of the underwater 4-channel digital device for detection of acoustic signals from high energy neutrinos.
A Master trigger signal from NT200+ Neutrino Telescope
Antenna A device container
ARTICLE IN PRESS S132
V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
throughout the whole water column. The bottom of a water basin can represent a source of seismic noise and noise associated with gas seeps. Thermal noise represents the lower bound of underwater noise and determines the minimum threshold of acoustic neutrino detection [23]. All the spectra obtained from our measurements can be classed into two groups. The first group includes the spectra with power spectral density (PSD) of noise that decreases more or less uniformly with increasing frequency in the frequency band under study (20–40 kHz). Fig. 4 shows examples of the noise PSD that belong to the first group. This type of noise is produced by multiple sources uniformly distributed near the surface, and is observed under stable and calm meteorological conditions. The qualitative behaviour of noise PSD in Lake Baikal does not contradict the empirical frequency dependence typical for ocean sites [22] with a slope of spectra of about 5–6 dB/octave on average. In winter, the spectra are usually somewhat steeper than in summer. The spectra, which contain some specific features in certain frequency bands, belong to the second group. These spectra are observed under nonstationary, nonuniform meteorological
-100
AS1 AS2 AS3 AS4
Sensitivity, [dB re V/uPa]
-102 -104 -106 -108 -110 -112 -114 -116 -118 -120 15
20
25
30
35
40
Frequency, [KHz] Fig. 3. Sensitivity of acoustic sensors (including 70 dB pre-amplifier gain) in units of dB V/mPa as a function of frequency.
Fig. 2. Acoustic sensor with housed pre-amplifier and hydrophone H2020C.
4. Acoustic noise characteristics at Baikal site Ambient noise in the ocean has been much investigated [21,22]. The most intense sources of noise are waves, wind, rain, ice cracks, ship traffic, etc., which are located in the surface water layer; biological noise and thermal noise are generated
Fig. 4. Examples of the power spectral density of noise (PSD) that belong to the first group (see text).
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
conditions, and are due to ship traffic, gas seeps, and so on [24]. Spectral features may also arise in the presence of some noise sources with specific spectra not far from the measurement site. Fig. 5 shows a data sample of the acoustic noise level in the 20–40 kHz interval, recorded by the 4-channel device. Periods of nonstationary conditions are visible by their high noise contribution. We also note that, in the measurements performed under stable meteorological conditions and in the absence of quasi-local noise sources, we do not observe a relative decrease in the highfrequency spectral components with increasing depth, which is observed in salt water (see, e.g., Ref. [15]). The latter fact is explained by the weaker attenuation of high-frequency waves in fresh water. At any point of time the noise depends on the surface condition rather than on depth. Fig. 6 shows the smoothed time dependences of the acoustic noise amplitude for several depths in the frequency band 22–44 kHz for submersions from the ice cover (several independent measurements in March 2004). Fig. 7 shows RMS of noise distribution. It is seen that at stationary and homogeneous meteorological conditions the
S133
integral noise power in the relevant frequency range 20–40 kHz reaches levels as low as 1 mPa (one of the lowest levels measured at currently considered acoustic neutrino detector sites). The mean value of the acoustic noise is around 2.5 mPa.
5. Pulsed noise characteristics From the assumption that the main mechanism of acoustic signal generation by cascade showers is thermoacoustic [5], it follows that the pressure pulses produced by them should have a bipolar shape. The sign of the first half-wave of a pulse is positive if the water temperature at the site of shower development is higher than the temperature corresponding to maximum water density, TMD, and the heating of water by the cascade leads to its expansion. In the opposite case, the sign of the first half-wave is negative.
Fig. 7. Noise probability (20–40 kHz), as measured with the device in 2006.
5 Fig. 5. Acoustic noise level in the frequency band 20–40 kHz at the depth of 150 m.
4.5
Temperature Profile, 20080321 Temperature of maximal density
4
D = 0050[M] D = 0100[M] D = 0300[M] D = 0500[M] D = 0700[M] D = 0900[M] D = 1100[M]
-30 -35
Temperature, C
RMS of acoustic noise [dB re Pa]
-25
10 mPa
-40 -45
3.5
3
2.5
-50 2
-55 1 mPa
-60
1.5
-65 15
20
25
30
35
40
45
Time [hours]
1
0
200
400
600
800
1000
1200
1400
Depth, m Fig. 6. Time dependence of the acoustic noise in the frequency band 22–44 kHz and depth range 50–1100 m at period of 23–30 March 2004 (local time since midnight of first measurement day).
Fig. 8. Measured temperature as a function of depth (28.3.2008) and temperature of maximal density, TMD (dashed line, calculated from Ref. [26]).
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
The Baikal water temperature at depths below 400 m is very stable around 3:4–3:6 C, see Fig. 8 for a spring 2008 measurement. As also shown in Fig. 8, the temperature is only equal to that of maximal density (TMD—dashed curve calculated using the formula from Ref. [26]) at shallow depths below 200 m; at larger depths they differ significantly, since TMD falls steeply by 0:2 C per 100 m depth. Due to the elongated shape of the showers the acoustic signals are emitted in a pancake-shaped volume with a few degrees opening angle. The expected pulse duration is on the order of several tens of microseconds and is proportional to the shower diameter. The pulse amplitude is proportional to the released energy density. Hence, for determining the properties of acoustic noise as the background against which neutrino induced cascades should be detected, it is necessary to measure the amplitude–time characteristics of this noise in the high-frequency spectral region. Several years of study of acoustic noise in Lake Baikal have shown that the background for acoustic detection of high-energy showers is primarily represented by pulsed noise with short pulse duration [24,25]. The acoustic antenna allows us to estimate the zenith and azimuth angles of incidence of acoustic signals. The position of antenna’s hydrophones in water is fixed by acoustic transponders located around NT200+. Fig. 9 presents an event with bipolar pulses in all four channels, detected by this device. Fig. 10 shows the zenith angle distribution for events with bipolar pulses registered in April–May 2006. Most pulses are located in the vicinity of the horizontal plane (zenith 90 ). Sources of pulses coming from just below horizon are likely located also in the near-surface zone. They appear to come from below horizon due to refraction which is caused by the growth in sound velocity with depth. From the region within 45 around the opposite zenith, i.e. from the deep lake zone below the device, no events with bipolar pulse form have been observed. Fig. 11 presents a comparison between amplitudes for showers of given energy expected at distances of 100 and 1000 m in the Mediterranean Sea and in Lake Baikal, calculated using the formula from Refs. [7,26,27]. Taking into account that the highfrequency noise level in Lake Baikal can be as low as 1 mPa (horizontal line at Fig. 4), it is possible to register showers of energy above 1018 eV from a distance of 100 m, and above 1019 eV
Fig. 10. The distribution of reconstructed zenith angle for events with bipolar pulses.
1 Lake Baikal Mediterranean Sea 100 m 1000 m
10-1 Signal amplitude, Pa
S134
10-2 10-3 10-4 10-5 10-6
1015
1016
1017
1018
1019
1020
1021
E, eV Fig. 11. Dependence of expected acoustic signals amplitude generated by highenergy showers versus shower energy.
from 1000 m. At distances much larger than 1 km, signal amplitudes of showers in Baikal water are similar or higher than in the Mediterranean due to lower sound absorption in Baikal freshwater.
6. Summary
Fig. 9. A triggered event with bipolar pulses (channels 1–4).
A prototype device for detection of acoustic signals from ultrahigh-energy neutrinos in water was constructed and is operating since April 2006 at depth 150 m at the site of the Baikal neutrino telescope. Our measurements show that the integral noise power in the frequency band 20–40 kHz can reach levels as low as about 1 mPa. The main source of the noise including bipolar pulses is the near-surface zone of the lake. Analyzing the sound wave arrival directions, we did not find any bipolar pulses which were generated by sources located at large depth of the lake. Thus, we suggest an acoustic detector design that is based on the deployment of a grid of rather compact acoustic antennas arranged at shallow depths (100–200 m for Lake Baikal), which
ARTICLE IN PRESS V. Aynutdinov et al. / Nuclear Instruments and Methods in Physics Research A 604 (2009) S130–S135
mainly ‘‘watches’’ the water volume from the surface down to the bottom. We conclude from the ongoing work that favourable conditions for acoustic neutrino detection at Lake Baikal exist clearly motivating further activities.
Acknowledgements This work was supported by the Russian Ministry of Education and Science, the German Ministry of Education and Research, REC-17 BAIKAL and the Russian Fund of Basic Research (Grants 0702-00791, 08-02-00198, 08-02-10001, 08-02-05037, and 08-0205037), and by the Grant of President of Russia NSh-4580.2006.2 and by NATO-Grant NIG-9811707 (2005), and in part by the program ‘‘Development of Scientific Potential in Higher Schools’’ (projects 2.2.1.1/1483, 2.1.1/1539, 2.2.1.1/5901). References [1] V. Aynutdinov, et al., Astropart. Phys. 25 (2006) 140. [2] V. Aynutdinov, et al., Nucl. Instr. and Meth. A 588 (2008) 99 V. Aynutdinov, et al., astro-ph/0811.1109, Nucl. Instr. and Meth. A, in press. [3] A. Achterberg, et al., Astropart. Phys. 26 (2006) 155. [4] J. Carr, ANTARES Collaboration, Nucl. Instr. and Meth. A 588 (2008) 80.
S135
[5] G.A. Askaryan, Atomnaya Energiya 3 (1957) 152. [6] C.S. Clay, H. Medwin, Acoustical Oceanography, Wiley, New York, 1977. [7] G.A. Askaryan, B.A. Dolgoshein, A.N. Kalinovsky, Nucl. Instr. and Meth. A 164 (1979) 267. [8] J.G. Learned, Phys. Rev. D 19 (1979) 3293. [9] G.A. Askaryan, B.A. Dolgoshein, Preprint no. 160, Lebedev Physical Institute of the USSR Academy of Sciences, Moscow (1976); G.A. Askaryan, B.A. Dolgoshein, Pis’ma ZhETF 25 (1977) 232. [10] R. Nahnhauer, S. Bo¨ser (Eds.), in: Proceedings of International Workshop on Acoustic and Radio EeV Neutrino Detection Activities (ARENA2005), World Scientific, Singapore, 2006, p. 298. [11] L.F. Thompson, Nucl. Instr. and Meth. A 588 (2008) 155. [12] S. Danaher, ACoRNE Collaboration, J. Phys. Conf. Ser. 81 (2007) 012011. [13] K. Graf, et al., J. Phys. Conf. Ser. 81 (2007) 012012. [14] G. Riccobene, NEMO Collaboration, J. Phys. Conf. Ser. 81 (2007) 012013. [15] N. Kurahashi, et al., Int. J. Mod. Phys. A 21S1 (2006) 217. [16] S. Boeser, et al., Int. J. Mod. Phys. A 21S1 (2006) 221. [17] V. Aynutdinov, et al., Phys. Atom. Nuclei V 69 (11) (2006) 1914. [18] V. Aynutdinov, et al., Nucl. Instr. and Meth. A 567 (2006) 433. [19] V. Aynutdinov, et al., in: Proceedings of 29th International Cosmic Ray Conference, Pune, India, August 3–10, 2005, astro-ph/0507715. [20] V.M. Aynutdinov, et al., in: Proceedings of the 29th ICRC, vol. 8, Pune, India, 2005, p. 251. [21] V.O. Knudsen, R.S. Alford, J.W. Emling, J. Mar. Res. 7 (1948) 410. [22] A.V. Furduev, in: Ocean Acoustics, Nauka, Moscow, 1974, pp. 615–691 (in Russian). [23] R. Mellen, J. Acoust. Soc. Am. 25 (1952) 478. [24] V. Aynutdinov, et al., Acoust. Phys. 52 (2006) 495. [25] V. Aynutdinov, et al., Int. J. Mod. Phys. A 21S1 (2006) 117. [26] C.T. Chen, F.J. Millero, Limnol. Oceanogr. 31 (1986) 657. [27] R.F. Weiss, E.C. Carmac, V.M. Koropalov, Nature 349 (1991) 665.