The residual of noise rate is the difference between the sum of the noise rate and the moving average of the sum. Here is showed the distribution of residual of noise rate for the arrays AMANDA B4 and B10 without any cuts, and the contour plots of the fitted Likelihood function versus the residual of noise rate again for AMANDA B4 and B10 without any cuts. The variables of the distribution have no weights.

The distributions with the cut on the reduced fitted Likelihood function > 1.5, keeping 99.9% of the signal. The scatter plots for the same cut. In addition are showed also the cut values for a detection efficiency at 90% for a Supernova like SN1987A at 8.5 and 6 kpc.

The DELTA_MU is the best value of the Maximum Likelihood estimate, this shows the corresponding increasing value of the noise rate for each PMT, by the occurrence of a SN close to the Galactic center. The scatter plots of the DELTA_MU versus the residual of the noise rate for AMANDA B4 and B10 without any cuts.

The Comparison between the data and the Monte Carlo for the scatter plots DELTA_MU vs. Residual of the noise rate. Here are displayed the plots with the cut of the fitted Likelihood < 1.5, on the MC are showed a hypothetical SN close to the galactic center.

The plots show the distribution of the best value of the Maximum Likelihood estimate, with and without any cuts. The distributions in this case are weighted concerning the sigma of the noise rate Gaussian distribution for each PMT. Interesting is to note that the sigma of the distribution for AMANDA B10 is smaller than AMANDA B4. Preliminary the DELTA_MU values are calculated as integer value.

The simulated MC distributions show also a better behaviour of the array B10 compared to the B4. In addition are displayed a hypothetical increasing of the value DELTA_MU due to SN signal close to the Galactic center.

The AMANDA Detector is sensitive only inside our Galaxy. The picture shows a simulation of the Milky Way following the variation in star density distribution according to the Bahcall & Soneira model.

Here are showed the distribution of the fraction of stars as function of the distance starting at the position of the Sun. Together with the Bahcall's model is displayed also the Monte Carlo distribution, which has a sharper increasing close to the center of the galaxy. Instead of as fraction of stars one can observe the distribution as number of the Supernova per century vs. distance in our Galaxy.

The current sensitivity of AMANDA B4 is close at 7.5 kpc with a ratio S/N at 1. The upgrade of this array feature with 250, 1000, 5000 PMTs are showed. Interesting is to reach the distance at 24 kpc with S/N at 1 through an array designed with 5000 PMTs.

The main goal is to be sensitive around at 95% of the SN rate in our Galaxy, but this result depends strongly on the number and features of the PMTs. The plots display the S/N for PMTs with a noise rate at 300 Hz and PMTs with a corresponding noise rate at 1160 Hz. Further is also showed the different values between the Bahcall's model and the MC expectation.