Figure 20: The AMANDA strings deployed at the South Pole. In the center bottom portion of the figure is the AMANDA-B10 array; the data from this array taken in austral winter 1997 is the subject of the analysis discussed here.
Back to proposal index
Previous section
Bibliography 
Our estimates of the performance of the IceCube detector are based on experience with AMANDA-B10 (taking data since 1997) and AMANDA II (soon to be completed). We discuss the result from the analysis of the first year of data, compare with simulations, and show how these extrapolate to IceCube.
Calibration of the first-generation AMANDA detector was completed in August 98. Although preliminary data were presented at the winter conferences, analysis of the first year of data has been completed. We have published the observation of atmospheric neutrinos with the partially deployed 1995–1996 detector which consisted of only eighty photomultiplier tubes. Manuscripts exist covering high energy atmospheric neutrinos and the search for neutrinos from active galaxies, cold dark matter particles, relativistic magnetic monopoles, and neutrinos from supernova collapses. Some of the results have been discussed in the science section of the proposal.
Figure 20: The AMANDA strings deployed at the South Pole. In the center bottom portion of the figure is the AMANDA-B10 array; the data from this array taken in austral winter 1997 is the subject of the analysis discussed here.
As predicted from transparency measurements performed with strings near 1 km depth, it was found that ice is bubble-free below 1400 m. Calibration of the detector (optical properties of the ice, geometry of the detector, cable time-delays) was completed in the austral summer 97–98. We found that
Figure 21: Deviation of the summed noise rate from the moving average for 51 selected PMTs in AMANDA during 108 live days in 1997, with a cut in chi2/NDF less than or equal to 1.5. Indicated are the cut-values for signals expected for SN1987A-type supernovae at 6kpc and 8.5 kpc.
After completion of the AMANDA detector with 300 OMs, a similar analysis established AMANDA as a neutrino telescope by observing atmospheric neutrinos at a rate consistent with expectation. Atmospheric neutrinos are identified using quality cuts on the reconstruction of muon tracks. We require a good likelihood of a Cherenkov cone fit and a significant number of "direct" photons smoothly distributed along a track. At high cut levels, typically more than 5 "direct" photons and muon range more than 100 m, all features of the neutrino events are consistent with expectations from Monte Carlo simulation of the atmospheric neutrino flux. For instance, the distributions of zenith angle can be seen in Figure 22.
Figure 22: Events remaining after strong cuts: muon track longer than 100 m, and number of "direct" photons greater than or equal to 6. The event rate and zenith angle distribution are consistent with the response of the detector to atmospheric neutrinos.
The neutrino event in Figure 23a, taken from the on-line event display, is particularly easy to understand. A muon generated in a neutrino interaction below the detector travels along a string at an average distance of 10 m, generating OM signals over a track length of 400 m. The arrival time of the first photon at an OM is plotted against the distance of that OM from the bottom of the string, reproducing the speed of light; see Figure 23b. Because the muon track is misaligned with the string by 1.5°, the projection of the track on the string travels with a speed somewhat less than c as shown in the figure. Note that the arrival times of the photons on a second string, although distant by 30 m from the muon track, still indicate its direction despite the scattering of the photons in the ice.
Figure 23: (a) Neutrino event in AMANDA. The shading of the dots represents time of triggered photomultipliers, and the size represents amplitude. The reconstructed muon track moves upward over 400 m. (b) Arrival times of the first photon as a function of distance along the two strings for the event in (a).
Another neutrino event is shown in Figure 24, which looks strikingly similar to a simulated upgoing neutrino event in the same direction. By requiring the long muon track the events are gold-plated, but the threshold is rather high, roughly Enu greater than or equal to 50 GeV. This type of analysis now allows AMANDA to harvest roughly one high energy atmospheric neutrino per day, adequate for calibration of the detector. It is impressive that three analyses with two independent sets of software tools are able to extract largely overlapping neutrino samples from the data.
Figure 24: (a) Neutrino event in AMANDA. (b) Simulation of an upgoing muon track with the same direction as the observed event in (a).
Figure 25: Skyplot from point source search analysis.
The potential of AMANDA as a dark matter detector relies on observing the sun, as previously discussed. This has to await the completion of AMANDA II because it requires good sensitivity near the horizon. In the meantime we have performed an analysis searching for the annihilation of WIMPs in the core of the earth. Figure 26 compares the present upper limit with other experiments.
Figure 26: Comparison of AMANDA limits on WIMP masses with those from other experiments.
AMANDA allows us to put stringent limits on relativistic Dirac monopoles. These exotic particles would have a spectacular signature produced by their highly ionizing radiation. As a result, a detector of AMANDA's size does not need a large data set to improve on existing upper limits: Figure 27 shows a comparison with other experiments.
Figure 27: Comparison of AMANDA limits on relativistic monopoles with those from other experiments.
Finally, the seasonal variation of muon flux demonstrates an understanding of the detector response. The muons originate mostly from the decay of cosmic ray interaction products such as pions and kaons. The lower the air temperature, the denser the air between where the pions and kaons are produced and where they decay, the more of them interact before they can decay, and the fewer the muons that reach AMANDA. Figure 28 shows the changes in muon flux closely following the changes in temperature, just as expected.
Figure 28: Percentage change in downgoing muon flux, and percentage change in temperature over time.
return to top of page