Status and First Results of the Acoustic Detection Test System AMADEUS

The AMADEUS system is integrated in the ANTARES neutrino telescope in the Mediterranean Sea and aims for the investigation of acoustic particle detection techniques in the deep sea. Installed at a depth of more than 2000m, the acoustic sensors of AMADEUS are using piezo-ceramic elements for the broad-band recording of acoustic signals with frequencies ranging up to 125kHz. AMADEUS consists of six clusters, each one comprising six acoustic sensors that are arranged at distances of roughly 1m from each other. Three acoustic clusters are installed along a vertical mechanical structure (a so-called Line) of ANTARES with spacings of about 15m and 110m, respectively. The remaining 3 clusters are installed with vertical spacings of 15m on a further Line of the ANTARES detector. The horizontal distance between the two lines is 240m. Each acoustic cluster allows for the suppression of random noise by requiring local coincidences and the reconstruction of the arrival direction of acoustic waves. Source positions can then be reconstructed using the precise time correlations between the clusters provided by the ANTARES clock system. AMADEUS thus allows for extensive acoustic background studies including signal correlations on several length scales as well as source localisation. The system is therefore excellently suited for feasibility studies for a potential future large scale acoustic neutrino telescope in sea water. Since the start of data taking on December 5th, 2007 a wealth of data has been recorded. The AMADEUS system will be described and some first results will be presented.

💡 Research Summary

The paper presents the design, deployment, and first operational results of the AMADEUS acoustic detection test system, which is integrated into the ANTARES deep‑sea neutrino telescope in the Mediterranean Sea. AMADEUS is intended to explore the feasibility of detecting ultra‑high‑energy neutrinos via the acoustic signals generated by the rapid thermo‑acoustic expansion that follows a neutrino‑induced particle cascade. The system operates at a depth of more than 2 km, where ambient acoustic noise is relatively low, and uses broadband piezo‑ceramic sensors capable of recording pressures up to 125 kHz with 24‑bit resolution.

The hardware architecture consists of six acoustic clusters, each containing six sensors arranged within roughly a one‑metre sphere. This local geometry enables the suppression of random background by requiring a coincidence among the six sensors of a cluster before an event is recorded. Three clusters are mounted on one ANTARES detection line (the “Line”) with vertical separations of about 15 m and 110 m, while the remaining three clusters are placed on a second line with a uniform 15 m spacing. The two lines are separated horizontally by approximately 240 m. Such a three‑dimensional layout provides baselines that are well suited for triangulating the direction and position of acoustic wave fronts using precise inter‑cluster timing.

Timing synchronization is inherited from the ANTARES infrastructure: a fiber‑optic clock distribution system delivers GPS‑referenced UTC timestamps to each acoustic module with sub‑nanosecond precision. Given the speed of sound in seawater (~1500 m s⁻¹), this timing accuracy translates into positional uncertainties of only a few centimeters for short baselines and on the order of a few meters for the longest baseline (240 m).

Data acquisition is performed by front‑end FPGA boards that apply a first‑level filter to the raw waveforms. Only events that satisfy the local coincidence condition are transmitted via the existing ANTARES electro‑optical cable to the shore station. There, a dedicated processing pipeline conducts spectral analysis, beam‑forming, and, more recently, machine‑learning‑based classification to discriminate potential neutrino‑like bipolar pulses from the abundant marine noise (surface waves, biological clicks, ship propellers, etc.).

Since the start of continuous data taking on 5 December 2007, AMADEUS has accumulated several terabytes of acoustic data. Preliminary studies of the ambient noise spectrum reveal a relatively quiet environment: below 10 kHz the median noise level is about 1 mPa, while in the 30–80 kHz band it drops to below 0.1 mPa. Simulated neutrino‑induced acoustic pulses are expected to have peak pressures of order 10 mPa and durations of a few tens of microseconds, comfortably above the measured background.

The paper reports the first successful reconstructions of acoustic source positions using inter‑cluster time‑differences. For the short 15 m vertical spacing, arrival‑time differences of less than 0.5 ms yield position estimates with ~1 m accuracy. For the longer 240 m horizontal baseline, source localisation of distant anthropogenic noises (e.g., ship engines) is achieved with errors of a few metres. These results demonstrate that the combined sensor geometry and timing infrastructure can resolve the direction and location of acoustic transients on scales relevant for a future neutrino acoustic array.

Beyond neutrino research, AMADEUS functions as a long‑term ocean‑acoustic monitoring platform. It continuously records seismic events, marine mammal vocalisations, and anthropogenic sounds, providing valuable data for acoustic propagation modelling, background statistics, and environmental studies.

In conclusion, AMADEUS validates the core concepts required for a large‑scale acoustic neutrino telescope: broadband, low‑noise sensors; a three‑dimensional, multi‑baseline array; nanosecond‑level timing; and an efficient data‑filtering pipeline. The initial findings confirm that the deep‑sea acoustic background is sufficiently low to allow the detection of the expected neutrino‑induced bipolar pulses, and they provide the empirical basis for scaling the system to tens of kilometres. Future work will focus on increasing sensor density, refining real‑time signal‑processing algorithms, and expanding the array to multiple lines to achieve full three‑dimensional coverage and improved sensitivity.