摘要:
Acoustic neutrino detection is a promising method to observe ultra high energy neutrinos. These neutrino with energies larger than 10 18 eV have a relatively low expected flux, thus a large instrumented volume is required. Current estimations give an instrumented volume of around O(100) km 3 using hydrophones as the detection modules. Measuring ultra high en- ergy neutrino would provide us with crucial information on extragalactic sources, the GZK cut-off, and also the CνB. In this work, an event detec- tion algorithm based on clique, a subspace clustering algorithm, was de- veloped. Furthermore, a first look at event reconstruction was taken. Ultra high energy neutrino events with energies of roughly 5 · 10 19 eV were ap- proximated by using their characteristic pancake shape. An instrumented volume of 4 km 3 was simulated. The study has shown that a hydrophone density of 400 per km 3 would provide the desired detection efficiency of near 100%. At these values a noise rate of 5 Hz can be suppressed using an amplitude criterion besides a causality one. A causality criterion alone suppresses a noise rate of 0.5-0.6 Hz. Furthermore, it was found that a con- figuration utilizing multiple detector blocks would maximize the effective volume of the detector. Moreover, the hydrophones should be designed for a sensitivity in the range of 0-15 kHz. The reconstruction algorithm tested did not provide the desired results, it is recommended to develop an algorithm specifically for acoustic neutrino detection. We found that design of the detector is a balancing act between detection efficiency, de- tector size, and noise suppression. Finally, this study demonstrates the possibility of using a clique based approach for event detection in ultra high energy neutrino detection. However, we recommend the develop- ment of noise suppression algorithms at the single waveform level, as sup- pression of a noise rate of 0.5-0.6 Hz or 5 Hz, depending on match criteria, is not enough based o