AMANDA

AMANDA uses photomultiplier tubes to detect the optical Cherenkov cones which result from muons produced in muon neutrino charged current interactions. Its architecture is optimised for reconstructing upgoing neutrino-induced muon of typical energy

10¹¹ - 10¹⁵ eV. By looking for upgoing muons it is possible to control the background associated with the muons produced by cosmic ray interactions in the earth's atmosphere. AMANDA is sensitive to atmospheric neutrinos produced in the northern hemisphere atmosphere of the earth as well as the higher energy neutrinos discussed in the introduction. In comparison the Japanese Super Kamiokande neutrino detector is limited by its size to detecting neutrinos with energies below 10¹⁰ eV.

During the 1993-94 austral summer, four strings were deployed with photomultiplier modules at depths of 800 to 1000m. However residual bubbles in the ice scattered the Cherenkov photons preventing a proper reconstruction of the muon path. A deeper array, AMANDA-B has been deployed in bubble-free ice below 1500m (at a depth of 1300m the last air bubbles have transformed into air hydrate crystals). There is still some scattering due to dust in the ice, however this is isolated to several well-defined depths (Woschnagg 1999) and the current array has enabled the reconstruction of muon paths. A coincidence trigger with the South Pole Air Shower Experiment is being used to calibrate pointing accuracy and ice properties (Miller 1999). AMANDA-B consists of 10 strings with 300 optical modules each consisting of a photomultiplier tube, presenting an effective area for muon tracking on the order of 10⁴m² (Hill 1999).

Analysis of 113 days of AMANDA-B data from April to November 1997 has yielded 16 neutrino candidate events. These 16 candidates were extracted from around 10⁸ events. About 90% of the cosmic ray muons are rejected with a simple filter method using the correlation of arrival times and depth of the observed Cherenkov photons. The remaining data are reconstructed by fitting the Cherenkov light cone generated by a relativistic particle to the observed arrival times. A set of quality cuts are used to further reduce the data. These cuts are based on the number of direct hits, the length of path that the hits are distributed over (at least 100m) and requiring that the event was not concentrated at the top or bottom of the detector. The characteristics of the observed neutrino candidates are in agreement with atmospheric neutrino Monte Carlo simulations (Karle 1999). A full simulation of atmospheric neutrinos predicts that 21 events pass the cuts described.

The most optimistic predictions for the diffuse neutrino flux predict an event rate of a few per year for AMANDA-B with other intensity estimates being marginal. During the 1997-98 season construction of AMANDA-II was begun, which will have an effective area several times larger than AMANDA-B. Three strings were deployed with optical modules ranging from 1300m to 2400m. These will be used to investigate the optical properties of polar ice over this depth range and will constitute the first three strings of the new detector. The geometry is shown in figure 1. Planning is underway to instrument a cubic volume of ice - the IceCube Observaory. The AMANDA collaboration anticipate of the order of 10 neutrinos per year will be detected from sources such as AGN and GRBs with a km² telescope (Halzen 1999).

The most promising area for the current AMANDA instrument is transient neutrino sources. The current array has a pointing accuracy of 2.5 degrees per muon track (Hill 1999). The Burst and Transient Source Experiment (BATSE) onboard the Compton Gamma Ray Observatory provides position and time for GRBs which means that several of the quality criteria cuts can be relaxed. The procedure used to search the data is reviewed by Bay (1999) and Kim (1999).

Figure 1: Schematic view of AMANDA-A, AMANDA-B and the three new strings. (Picture drawn by Alexander Biron and Thorsten Schmidt of DESY.)

© Copyright Astronomical Society of Australia 1997