Latest Results from the Milagro Observatory

1
Latest Results from the Milagro Observatory

• Vlasios Vasileiou
• NASA Goddard Space Flight Center University of
  Maryland, Baltimore County

2
Overview

• The Milagro Detector
• (Some of the) Latest Results
• Gamma-ray Observations
• Survey of the galactic plane
• Diffuse emission from the galactic plane
• Cosmic Rays
• Large-scale anistropy
• Discovery of two regions of excess Cosmic Rays
• Gamma-Ray Bursts
• Triggered untriggered (blind) searches
• Conclusion

3
The Milagro Observatory
4
The Milagro Observatory

• Water-Cherenkov detector located at the Jemez
  mountains near Los Alamos, New Mexico.
• Elevation 2630 m
• Detector Components
• Central Pond
• Outrigger Array

5
The Central Pond
Primary Particle
Thickness 1 m

• 24 Million liter reservoir of highly purified
  water
• Covered with a light-tight cover
• 80m x 60m x 8m (depth) (5000 m2)?
• 723 PMTs arranged in two layers
• Air Shower Layer 450 PMTs under 1.4 m of water
• Triggering
• Direction Reconstruction
• Muon Layer 273 PMTs under 6m of water
• Background Rejection
• Energy Reconstruction

?
Shower Front (Diameter 100 m)?
50 meters
80 meters
6
The Outrigger Array

• Outrigger Array
• 175 Water tanks spread over 40,000 m2
• Contain water and a downwards facing PMT
• Added 2003
• Improved
• Effective area
• Angular resolution
• Energy resolution
• Background rejection

7
Milagro's Performance

• Angular reconstruction accuracy 0.3o-1.4o
• Most of the effective area at TeV energies
• 105 m2 _at_ 10 TeV
• 10 m2 _at_ 100 GeV
• Median energy of triggers few TeV (for a
  Crab-like source)
• Performance
• Wide field of view (2 sr)
• High duty cycle (90)
• Good for unbiased whole-sky searches,
  observations of large-scale features
  anisotropies, monitoring for transient emissions
  (flares, GRBs).
• Crab-like source
• Milagro 8s/sqrt(year)

8
The TeV Sky as Seen by Milagro
9
The Northern Sky at 20 TeV
Crab Nebula

• 6.5 year data set (July 2000-January 2007)?

ApJ 664 (2007) L91
10
Cygnus Region
GeV 19070557

• GeV Sources

Geminga

• Milagro has discovered 3 new sources 4
  candidate sources in the Galaxy.
• 5/7 of these TeV sources have GeV counterparts
  (only 13 GeV counterparts in this region -
  excluding Crab)
• Probability 3x10-6

11
Diffuse Emission from the Galactic Plane
12
Source Exclusion

• The signal regions (shown with the squares) were
  fit with a two-dimensional Gaussian plus a
  constant.

13
Diffuse Emission
Cygnus Region with Matter Density Contours
overlaying Milagro Significance
Source Subtracted Longitude Profile by Milagro
GALPROP (optimized)? Sum po decay Inverse
Compton
Cygnus Region
Below Horizon
14
Milagro and GALPROP predictions

• Extragalactic diffuse
• Bremsstrahlung
• po decay
• Inverse Compton (dashed line IC on the CMB)
• Sum of the above three contributions

15
Profiles of the Galactic Diffuse Emission

• Inverse Compton component extends to higher
  latitudes
• Pion decay due to interactions with matter mostly
  at low latitudes
• Profile of Milagros measurements at Cygnus
  region is narrower than GALPROPs predictions -gt
  possibly a stronger pion component is involved

Cygnus Region 65oltlongitudelt85o

• Inner Galaxy
• 30oltlongitudelt65o

• GALPROP Model
• ?o decay Inverse Compton Total

16
Discovery of Two Regions of Excess Cosmic-Rays
17
Milagro Observes Anisotropy in 10 TeV Cosmic Rays

• Milagros standard point-source analysis with a
  10o bin size
• Results
• Two regions of fractional excess of 6e-4 (Region
  A) and 4e-4 (Region B) above the cosmic ray
  background were detected.
• Composition
• Excesses are not gamma rays (or electrons), but
  charged cosmic rays (8.6s Region A and 6.6s
  Region B). ?
• Energy Spectrum
• The spectra of both excesses are inconsistent
  with the cosmic-ray spectrum (4.6s and 2.5s)
• Spectrum of region A Broken power-law with index
  -1.45 and break energy9TeV.

Galactic Plane
Heliotail
Geminga
18
Tibet Collaboration ICRC Merida 2007
Cygnus region
Mrk421
Crab
?
Tibet icrc 02007
19
Origin/Explanation of the Excesses

• Composition
• Not photons or electrons
• Neutrons from a star? Unlikely -gt 10 TeV neutrons
  decay in 0.1pc -gt much closer than the nearest
  star.
• Gyroradius of a 10TeV proton in a 2µG magnetic
  field (estimate of the local Galactic field) is
  only 0.005pc (1000AU).
• Magnetic field must connect us to the source and
  be coherent out to it (gt100pc).
• Tips
• Connection to heliosphere? Region A coincides
  with the direction of the heliotail.
• The direction of both regions is nearly
  perpendicular to the expected Galactic magnetic
  field direction.
• Multiple explanations were proposed
• Salvati Sacco, astro-ph0802.2181
• Drury Aharonian, astro-ph0802.4403
• K. Munakata for M. Amenomori AIP Conf Proc Vol
  932, page 283

20
Large-Scale Cosmic-Ray Anisotropy
21
Large-Scale Cosmic Ray Anisotropy

• To study anisotropies with scale larger than 10o,
  an alternative method was used.
• Can detect effects down to the 10-4 10-3 level.
• Measures the fractional (not absolute) anisotropy
  in the RA direction.
• Median Energy 6 TeV

22
Large-Scale Cosmic-Ray Anisotropy
Galactic North Pole

• Define central-deficit region
• Dec 5o-35o RA 160o-210o
• Symmetric around minimum at RA188o
• Average fractional anisotropy -2.85 0.06
  0.08 x 10-3 (20s)
• Coincident with the Galactic North Pole

23
Properties of the Central-Deficit Region
07/2007 Solar minimum
07/2000 Solar maximum

• Mean anisotropy of central-deficit region
  increases with time ( factor of 2 / 7 yrs).
• Trend present in all energies
• Tibet found no evidence of fluctuation Mean
  anisotropy at 1997-2001 2001-2005
• Average value of the anisotropy depends on the
  energy -gt decreases at high energies

24
Possible Explanations

• Effects that can create anisotropies
• Large scale or local magnetic field
  configurations
• Effects of the heliosphere
• Anisotropy observed to energies up to 100TeV -gt
  CR of such high energies are not easily
  influenced by the heliosphere
• Diffusion of cosmic rays out of the Galactic halo
• Supported by the fact that the deficit is close
  to the North Galactic Pole
• Contribution of discrete sources (such as
  supernova remnants)
• Compton-Getting effect -gt dipole effect due to
  the motion of the solar system with respect to
  the CR plasma -gt increase in CR flux of the order
  of 0.1 in the direction of motion.
• Energy independent
• Observed effect different than the predicted
  effect.

25
Gamma-Ray Bursts
26
Milagros Searches for GRBs

• Triggered
• In coincidence with external triggers (ex. from
  Swift)
• Using the reconstructed events (Egt100 GeV)
• Using the individual PMT hit-rates (Elt100GeV)
• Published upper limits at ICRC Merida Santa-Fe
  GRB Conferences 2007
• Comprehensive paper with upper limits in
  preparation
• Untriggered (Blind)
• Search of all of the Milagro data in space,
  starting time, duration
• Simple binned search
• Careful calculation of the effective number of
  trials
• Optimization of the bin-size versus the duration
  under search

27
Sensitivity to GRBs (Untriggered Search)
Fluence Sensitivity
Minimum Detectable Redshift
Swift data N. Butler et al. ApJ 2007
28
Blind Search for GRBs

• Searched 4.6 years of Milagro data for bursts of
  VHE gamma rays of duration from 0.1msec to
  316sec.
• The search was also sensitive to other transient
  phenomena such as the last stages of primordial
  black hole evaporation.
• No significant events were detected.
• Upper limits on the prompt VHE emission from GRBs
  were set.

• Best post-trials probabilities found in each
  duration
• Need to multiply these probabilities with an
  extra factor of 41 to account for the number of
  independent durations searched.

29
Setting Upper limits on VHE emission from GRBs

• Next step Set upper limits on the VHE emission
  from GRBs
• A VHE emission model that predicts -ln(1-CL) GRB
  detections by this search is excluded at the CL
  level.
• 2.3 detections -gt exclude a model at the 90
  Confidence Level
• A simulation of the GRB population was created to
  estimate the number of GRB detections.
• The simulation reproduced the (Eiso, z, t90)
  distributions of GRBs detected by Swift.
• Using the Swift GRB rate and the relative FOVs
  of Milagro and Swift we can calculate the rate of
  (detectable by Swift) GRBs in Milagros FOV.
• Using Milagros sensitivity data we can calculate
  the number of GRB detections by Milagro.

30
Simulation Details

• Redshift distributions
• Detected distribution Intrinsic distribution x
  detector-related selection function
• Selection function from Swift minimum detectable
  peak flux
• Intrinsic redshift distribution
• Short GRBs
• Compact-binary merger scenario
• Delay between the creation of compact objects and
  their final merger t follows P(t)1/t
• Rate of compact-binary mergers at a redshift z
  calculated by integrating the Star Formation Rate
  from z back to the past weighted by P(t)
• Long GRBs
• Star Formation Rate x fractional mass under some
  metallicity limit
• Duration distribution
• By Swift-detected GRBs
• This simulation can only constrain the prompt
  emission from GRBs

31
Simulation Details

• Isotropic Energy distribution
• Intrinsic Eiso distribution x detector-specific
  selection function
• Derived effective detection threshold in terms of
  S/sqrt(t90) from Swift data
• Intrinsic Eiso distribution
• Power-law with index -1.45
• Universal jet profile model

32
Verification of the Simulation

• Fluence distribution of Swift-detected GRBs. Top
  Short GRBs, bottom long GRBs
• These distributions depend on all the parameters
  involved in the simulation
• Intrinsic Eiso, z, t90 distributions
• Selection functions minimum detectable
  peak-flux, minimum detectable S/sqrt(T90)
• Excellent agreement between the simulations
  predictions and Swifts data.
• Curves with good statistics are the simulation
  results (top solid, bottom dashed).

Swift data
Swift data
Swift data N. Butler et al. ApJ 2007
33
VHE Emission Model

• Simple model on the VHE emission from GRBs
• Not all GRBs have VHE emission
• For the GRBs that do have VHE emission
• Isotropic energy emitted in the 1keV-10MeV energy
  range is proportional to the isotropic energy
  emitted in the 40GeV-EVHE,max energy range (where
  X is a cutoff energy results given versus
  various EVHE,max).
• VHE emission on a power-law results given
  versus various spectral indices.
• Upper limit set on the proportionality constant R

34
Upper Limits on the Prompt VHE Emission by GRBs
Number of detected GRBs (by this search) versus
the ratio R
Upper limit on R at the 90 Confidence Level
35
Upper Limits on the Prompt VHE Emission by GRBs

• Upper limits on R (90 CL) versus different
  spectral indices, maximum emitted energy, and
  upper metallicity limits.
• These results are for the case that all GRBs have
  VHE emission.

36
Latest Results

• 2007
• Discovery of TeV Gamma-Ray Emission from the
  Cygnus Region of the Galaxy
• Astro-ph 0611691 ApJ 658 (2007) L33
• TeV Gamma-Ray Sources from a Survey of the
  Galactic Plane with Milagro
• arXiv 0705.0707 ApJ 664 (2007) L91
• Milagro Constraints of the Very High Energy
  Emission from Short Duration Gamma-Ray Bursts
• arXiv 0705.1554 ApJ 666 (2007) 361
• 2008
• Discovery of Localized Regions of Excess 10-TeV
  Cosmic Rays
• arXiv0801.3827
• A Measurement of the Spatial Distribution of
  Diffuse TeV Gamma-Ray Emission from the Galactic
  Plane with Milagro
• arXiv0805.0417 Accepted at ApJ
• The Large Scale Cosmic-Ray Anisotropy as Observed
  with Milagro
• arXiv0806.2293 Submitted to ApJ
• In preparation
• Results of the triggered and untriggered GRB
  searches
• Search for TeV Pulsation from the Crab and
  Geminga
• Energy Spectra of Selected Gamma-Ray Sources

HAWC
37
TeV ?-rays A New Window on the Sky
0.1 GeV

• Milagro 10 TeV gamma-ray

38
Background and Signal
39
Probability distribution
40
AGNs

• HAWC will obtain duty factors and notify
  multiwavelength observers of flaring AGN in real
  time.
• Milagro has observed 7yr lightcurve of Mrk 421
• HAWCs increased sensitivity would result in 10x
  smaller error bars and have similar error bars on
  hour time scale rather than 64 days

Milagro and XTE ASM 7 yr lightcurve of Mrk 421
(Smith et al. ICRC 2007)
ASM Flux cts/s
Milagro - Events/day
MJD - 50000