Results from the Milagro GammaRay Observatory

1
Results from the Milagro Gamma-Ray Observatory

• Erik Blaufuss
• University of Maryland
• 2nd VERITAS Symposium

2
The Milagro Collaboration
A. Shoup, and G.B. Yodh University of
California, Irvine T. DeYoung, L.A. Kelley T.
Mai, D.A. Williams University of California,
Santa Cruz B. Shen University of California,
Riverside D. Berley, E. Blaufuss, U. Braver,
J.A. Goodman, E. Hays, J.E. McEnery, D. Noyes,
A.J. Smith, G. Sullivan University of Maryland
J. Ryan University of New Hampshire A.I.
Mincer, and P. Nemethy, L. Fleysher,
R.Fleysher New York University R. Atkins, M.M.
Gonzalez, M. Wilson University of Wisconsin R.W.
Ellsworth George Mason University C. Sinnis, B.
Dingus, C. Hoffman, X. Xianwu Los Alamos National
Laboratory
3
The Milagro Detector

• Located in mountains near Los Alamos, NM
• At elevation of 8600 (2600m)
• 60m X 80m X 8m covered pond
• Water-Cherenkov EAS detector

• 100 coverage, 50 eff.
• Wide field of view
• High duty factor (90)
• Sensitivity above 100 GeV

4
The Milagro Detector

• 723 8" PMTs distributed in 2 layers
• Top layer (450 PMTs) _at_ 1.5m depth
• Bottom layer (273 PMTs) _at_ 7m depth
• Completed pond construction in 1999
• Currently collecting data at a rate of 2000
  Hz
• So far, weve collected 100 billion events.
• Angular reconstruction of shower done with top
  layer
• Background rejection performed with bottom
  layer
• Total data rate 3MB/s (100TB/y raw data rate)
• Array of 175 outrigger detectors recently
  finished.

5
Detecting the Air Shower in Water
6
Separating Gamma Showers from Hadron background
MC proton event
MC ?-ray event

• A typical proton shower
• Clumpy pulse height distribution
• Patches in bottom layer of detector (muons,
  hadrons)

• A typical ? shower
• Smooth, uniform pulse height distribution
• No clumps

7
X2 -The ?/hadron cut
X2 NB2
/ PEmaxbot
? MC
Optimal cut X22.5 Removes 93 of background
Preserves 50 of ?-ray events Increases
sensitivity to ?-ray sources by a factor of 1.7
Data
Hadron MC
8
Moon shadow and Angular Resolution
The shadow of the moon in cosmic rays is observed
to be 30?. The shape of the deficit
perpendicular to deflection is used as a measure
of the angular resolution of the detector. This
gives an angular resolution for protons of
0.98o. This is in agreement with simulation.
Same simulation of ?-ray showers have an
angular resolution of 0.7o.
9
Observation of the Crab Nebula
At Crab Position (RA83.63 deg , Dec22.0
deg) Cuts ON Background
Excess Significance NFIT20 17922478
17915678 6800 1.5 ? NFIT20, 2291429
2282021 9408 6.0 ?
X22.5 Data collected over 900 days of
running with the pond detector. Observe 10.5
events/day. This observation is in good
agreement with the measured Crab flux from
the Hegra and Whipple experiments.
10
Milagro Energy Sensitivity
Energy distribution of reconstructed events from
a crab-like spectrum. Median event energy is
3 TeV, but still reconstructing events near
100 GeV
Median energy as a function of zenith angle.
11
Search for Flaring Sources

• We collect, reconstruct and apply gamma-hadron
  separation to events
• Look for excesses on the sky
• Every 2 hours, signal and background skymaps are
  generated and significant excesses searched for
• 2 hour maps are summed to make daily, weekly,
  monthly and yearly skymaps.
• A web page listing detected hot spots is being
  constructed

12
Online sky survey
13
All sky results

• Crab Nebula detected 6 sigma
• Publication ready for submission (g/h)
• Mrk421 flare of early 2001 5 sigma
• Mrk421 most recent flare (observed)
• Flux limits on Costamante et al,
• likely TeV sources soon.
• All sky survey capabilities
• important.

14
Milagros sensitivity to Flaring Objects
Work by E.Hays

• Flux required to trigger a 5s detection as a
  function of time

15
Untriggered GRB search
-Search the overhead sky on 50 time scales from
100msec to 100s -Probability distributions for a
few time scales
16
A candidate event
Significance is 3.2 x 10-13 Duration 0.63
sec Expect to see 15/year like this from
background fluctuations due to the large
amount of over sampling done. Monitor
detector health and GRB search Email
notification for most improbable
17
Milagros Sensitivity to GRBs
18
Milagro Sensitivity with GLAST
For a 1 second observation, Milagro becomes more
sensitive than GLAST above 100 GeV.
For observations of the prompt phase of a GRB,
current and future high energy gamma-ray
instruments (GLAST and Milagro) are very
complementary. (Spectra 30 MeV-10TeV)
19
Improving Milagros Sensitivity

• Recently, a great deal of effort has been focused
  on increasing Milagros sensitivity to gamma ray
  showers, especially at lower energies.
• The addition of an outrigger array surrounding
  the central pond.
• Expect at least a factor of 2 increase in
  sensitivity
• New, upgraded trigger
• Allows intelligent triggering and rejects
  background at the trigger level.
• New DAQ computer system.
• Allows for more complex analyses to be run in
  real-time online.

20
The Outrigger Array

• Recently added an array of 175 outrigger tanks
• -1500 gallon tanks
• - 4.6m2 x 1m high
• Array covers an area of 40,000m2 around pond.
• Will help determine shower core location
• -Better angular resolution
• -Improved g/h separation
• Now integrating into data and reconstruction

21
The Outrigger Array
22
Decreasing the Trigger Threshold
The background from single muons prevented us
from lowering the muliplicity trigger threshold
to increase our area for low energy ?-ray
events. Air shower's can be reconstructed with
as few as 15-20 air shower layer hits.
23
Trigger Upgrade- Summer 2002
A new trigger system was developed using a custom
FADC to examine the risetime of the trigger
pulse noticing that - Single muon events
deposit a large quantity of light locally
that spreads over the pond at vlight c/n
c/1.3. (large risetime) - Hits from low energy
air showers occur almost simultaneously.
(small risetime) The new trigger is 50
efficient for triggering on 20 PMT AS hit ?-ray
induced air showers.
24
Conclusions

• Milagro continues to operate and search the sky
  for sources.
• Developed a successful gamma-hadron separation
  technique and observed the Crab, Mrk421 flares.
• Improved sensitivity to sources
• Outrigger array
• New trigger
• Improved reconstruction being developed.
• Will continue GRB and transient searches