YouRen Wang

1
Point Source Analysis of
Combined Data (2000 2001 2002)
in Comparison to (2000 2001)

• You-Ren Wang
• Paolo Desiati
• University of Wisconsin-Madison

Oct. 12. 2004, Uppsala University
1
2
Blinded Analysis of (20002002)
2
3
Optimized Cuts (2000 2001) vs. (2000
2002)
declination bin radius nch
ldirb likeli. ratio
sensitivity -5.0 3.8 2.8
56.0 56.0 108.0 120.0 25.0
25.0 6.2726 4.8820 0.0
4.0 4.0 30.0 46.0 96.0
108.0 31.0 31.0 1.9255 1.4341
5.0 4.4 4.2 32.0
40.0 120.0 120.0 35.0 35.0
1.1772 0.87223 10.0 4.4
4.0 32.0 32.0 72.0 108.0
39.0 39.0 1.1628 0.84974 15.0
4.2 4.4 32.0 32.0
84.0 108.0 39.0 43.0 1.0821
0.84317 20.0 4.4 4.4
34.0 34.0 72.0 96.0 41.0
41.0 1.0796 0.82397 25.0
4.8 4.4 36.0 38.0 72.0
96.0 45.0 45.0 1.0882
0.79984 30.0 5.0 4.2
38.0 34.0 84.0 120.0 43.0
45.0 1.2084 0.88578 35.0
4.4 4.4 38.0 38.0 108.0
108.0 43.0 45.0 1.1789
0.87626 40.0 4.4 4.4
38.0 38.0 120.0 108.0 45.0 47.0
1.1969 0.86304 45.0
4.0 3.6 40.0 38.0 120.0 120.0
45.0 45.0 1.2146 0.83355 50.0
2.8 2.8 36.0 36.0
120.0 120.0 45.0 45.0 1.2631
0.84910 55.0 3.8 3.4
40.0 42.0 84.0 144.0 45.0
53.0 1.2598 0.94834 60.0
3.4 2.8 40.0 40.0 144.0
144.0 41.0 43.0 1.4404
0.99536 65.0 2.6 2.8
36.0 40.0 96.0 120.0 47.0
47.0 1.4801 1.0588 70.0
3.4 3.2 40.0 46.0 120.0 168.0
47.0 47.0 1.3136 0.95949 75.0
2.8 2.4 36.0 36.0
96.0 96.0 47.0 51.0 1.3248
0.93779 80.0 3.2 2.8
42.0 44.0 156.0 168.0 41.0 53.0
1.4001 0.95784 85.0
3.6 2.8 34.0 36.0 192.0 192.0
37.0 37.0 1.9011 1.2510
3
4
Number of Events after Final Cuts
Period
Region
4
5
Sensitivity
10-7 GeV cm-2 s-1
3 years
2 years
declination
5
6
Declination Averaged Flux (after Final Cuts)
event / solid angle
event / solid angle
event / solid angle
3 years
2 years
declination
declination
declination
Exp. Data
E-2 NUSIM
Atmos. NUSIM
6
7
Number Agreement between Data Atmos. NUSIM
after Final Cuts
2 years norm. factor 0.96
3 years norm. factor 1.03
7
8
Quality Agreement between Data Atmos. NUSIM
after Final Cuts
event
event
event
event
all zenith
all zenith
all zenith
all zenith
event
event
event
event
zenith 90110
zenith 90110
all zenith
all zenith
3 years
2 years
8
9
Median Neutrino Pointing Resolution ( Medium
Value of Delang Distribution of E-2 NUSIM)
degree
3 years
2 years
zenith
9
10
Effect of Nch Cut in Energy Selection (3 yrs
final cuts)
Solid w/o nch cut Dashed w/ nch cut
E-2 NUSIM
Atmos. NUSIM
logarithmic
GeV
GeV
linear
GeV
GeV
10
11
Unblinded Results of (20002002)
11
12
Sky Maps
2 years 959 events
3 years 1403 events
12
13
Full Sky Binned Grid Search (prepared
by Paolo)

• Sky divided in a number of bins
• Corresponding to optimized search radius vs
  declination
• With constant solid angle over the sky
• Grid shifted 4 times
• Statistical penalties accounted with toy MC
• 2000, 2001, 2002 (single years)
• Sky divided in 240-260 bins
• 2000-02
• Sky divided in 280 300 bins (bigger search
  radia)

13
14
Grid Search Results
14
15
Grid Search Results
15
16
Grid Search Results

• Binned search cannot see within the search bins

16
17
Grid Search Results
2000-01
2002
17
18
Grid Search Results

• In previous publication
• nobs 8
• nbkg 2.1
• 7/8 events with ltnchgt 30
• Optimized cut ltnchgt 35
• Events lost due to nch cut

• This is the highest fluctuation in the present
  2000 data (7 / 0.8)
• In previous publication 6 events were the same (7
  / 2.35)
• ?max 1.98 , p-value 1
• ltnchgt 49
• It was not significant

18
19
Limits of Selected Individual Sources
2 years 3 years
Candidate Dec. o RA h nobs
/nexp F? nobs /nexp F?
Markarian 421 38.2 11.07
0/0.7 0.7 0/1.4
0.3 Markarian 501 39.8 16.90
0/0.7 0.7 3/1.3 1.5 SS433
5.0 19.20
1/1.7  0.9 1/2.2 0.5 Cygnus
X3 41.0 20.54 0/0.7  
0.7 1/1.3 0.8 Crab Nebula
22.0 5.58 3/1.1   2.1
4/1.9 1.4 Cassiopeia A 58.5
23.39 2/1.0  2.1
2/1.1 1.4 M 87 12.4
12.51 2/1.1  1.6
3/1.8 1.2
19
20
Full Sky Flux Upper Limits
Dec o
Dec o
RA h
RA h
2 years
3 years
20
21
Cluster Analysis
number of occurance
3 years (607 events gt 8 dec)
2 years (378 events gt 8 dec)
0.05
probability of observing the same or higher
number of doublets in the toy MC
0.27
31
6
35
56
2
38
16
87
number of neighbors
Multiplet Distribution
a neighbor another event within the optimal
radius of an event (here 3.4 used)
10,000 sets of toy MC were generated, each set
has the same number of events (378 607)
for obtaining the distribution of multiplets.
Comparing exp. data multiplets with toy MC
multiplet distribution gives the probability
of observing the experimental number of multiplets
21