Combined Analysis of Cosmic-Ray Anisotropy with IceCube and HAWC - - PowerPoint PPT Presentation

Universidad de Guadalajara Centro Universitario de los Valles

Juan Carlos Díaz Véleza,b,

• M. Ahlersc, D. Fiorinod, P. Desiatib

Combined Analysis of Cosmic-Ray Anisotropy with IceCube and HAWC

Gamma-Ray Observatory Gamma-Ray Observatory

ICRC 2017 Busan, South Korea July 19, 2017

aCentro Universitario de los Valles, Universidad de Guadalajara, Guadalajara, Jalisco, México bWisconsin IceCube Particle Astrophysics Center (WIPAC) and Department of Physics, University of

Wisconsin–Madison, Madison, WI 53706, USA

cNiels Bohr Institute, University of Copenhagen, Copenhagen, Denmark

dDepartment of Physics, University of Maryland, College Park, MD, USA

Universidad de Guadalajara Centro Universitario de los Valles

2

Over the last few decades, several studies have measured a large scale anisotropy at 10‐3 level and a small-scale structure with an amplitude of 10−4 and angular size from 10° to 30°.

Large-scale features in South appear to be a continuation of those observed in the Northern Hemisphere.

HAWC South North Milagro ARGO-YBJ

Super-K

Tibet Array

Guillian et al., Phys Rev D, Vol 75, 063002 (2007)

Amenomori et al., Science Vol. 314, pp. 439 (2006)

Abdo et al., ApJ, Vol 698-2, pag 2121 (2009)

Zhang et al. Proc. 32nd ICRC (2009)

• A. U. Abeysekara et al. Astrophys. J. (2014)

IceCube

• M. G. Aartsen et al. Astrophys.J. 826 (2016)

Universidad de Guadalajara Centro Universitario de los Valles

Origin of anisotropy

3 Diffusive propagation from distribution of cosmic ray sources

Distribution of CR sources

A number of theories have proposed scenarios where the large-scale anisotropy results from the distribution of cosmic ray sources in the Galaxy and of their diffusive propagation

• P. Blasi and E. Amato arxiv/1105.4529

(a)

A.D. Erlykin and A.W.Wolfendale

• Astropart. Phys. 25 (2006) 183–194
• A. D. Erlykin and A. W. Wolfendale, Astropart. Phys. 25 (2006) 183–194.
• P. Blasi and E. Amato, JCAP 1 (2012) 11.
• V. Ptuskin, Astropart. Phys. 39 (2012) 44–51.
• M. Pohl and D. Eichler, Astrophys. J. 766 (2013) 4.
• L. G. Sveshnikova, et al. Astropart. Phys. 50 (2013) 33–46.
• R. Kumar and D. Eichler, Astrophys. J. 785 (2014) 129.
• P. Mertsch and S. Funk, Phys. Rev. Lett. 114 (2015) 021101.


Universidad de Guadalajara Centro Universitario de los Valles

Origin of small-scale anisotropy

4

Giacinti & Sigl arxiv/1111.2536

Propagation effects in ISMF

CR propagation Small-scale structure Turbulent GMF

Heliospheric effects

Desiati & Lazarian arxiv/1111.3075

Ripples in heliospheric boundary

CRs streaming along LIMF

CR scattering on ripples in the heliosphere boundary induce small-scale anisotropy.

It is expected that cosmic rays should lose any correlation with their original direction due to diffusion as they traverse through interstellar magnetic fields.

Giacinti & Sigl arxiv/1111.2536

+360 0.94 0.96 0.98 1 1.02 1.04 1.06 1.08

• 0.2
• 0.1

0.1 8.3 8.4 8.5 8.6 8.7 154 156 163 164 X (kpc) Y (kpc)

Giancinti & Sigl, Phys. Rev. Lett. 109, 071101 (2012) López-Barquero, et al, ApJ 830 19 (2016) Desiati and Lazarian, ApJ, 762, 1 (2013) Pogorelov et al., ApJ772 (2013) 2


• M. Zhang, et al, Astrophys. J. 790 (2014) 5

López-Barquero, et al ApJ, 842, 54 (2017)

Universidad de Guadalajara Centro Universitario de los Valles

Data Sets

IceCube HAWC

Hemisphere Southern Northern Latitude

• 90◦

19◦

Detection method

muons produced by CR

air showers produced by CR and γ

Field of view

• 90◦/-20◦, ∼4 sr (same sky over 24h)
• 30◦ /64◦ , ∼2 sr (8 sr observed)/24 h

Livetime 5 years 269 days over a period of 336.36 days Detector trigger rate 2.5 kHz 25 kHz

quality cuts energy cuts quality cuts energy cuts

Median primary energy 20 TeV 10 TeV 2 TeV 10 TeV Energy resolution (logE/GeV)

0.5 0.5 0.2 0.2

• Approx. angular resolution

2-3◦ 2-6◦ 0.3-1.5◦ 0.3-1.5◦

Number of events

2.8 × 1011 1.7 × 1011 2.6 × 1010 4.4 × 109

Data selected for analysis come from 5 years of the complete IceCube array, as well as 1 year of HAWC in its final configuration of 300 tanks. Only continuous sidereal days* of data were chosen for these analyses in order to reduce the bias of uneven exposure along right ascension.

5

Individual experiments have provided partial sky coverage that limits the interpretation

• f the results. This first full-sky combined observation at the same energy is done with

two observatories covering most of the celestial sphere.

* Gaps of 20 min. allowed within each 24 h period

Universidad de Guadalajara Centro Universitario de los Valles

Distribution of events as as a function of declination for IceCube and HAWC. Restricting datasets to overlapping energy bins significantly reduces statistics for HAWC. After cuts, both CR data sets have a median energy of approximately 10 TeV with little dependence

• n zenith angle.

6

• PRELIMINARY

log(E/GeV)

2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040

HAWC IceCube

PRELIMINARY

Universidad de Guadalajara Centro Universitario de los Valles

Iterative maximum likelihood method

Ahlers, BenZvi, Desiati, Díaz-Vélez, Fiorino, Westerhoff (arXiv:1601.07877)

7

• Traditional time-integration methods can strongly attenuate large-scale

structures exceeding the size of the instantaneous field of view for detectors located at middle latitudes.

• A fixed position on the celestial sphere is
• nly observable over a relatively short

period every day. The total number of cosmic ray events from a fixed position can only be compared against reference data observed during the same period.

• j q

j q

• ò

j q W = f a d f a d = d =

• a

d a d d = j q j q q ¢ =

• w

w w w w w =

• F -

F F

• F

F F DW

t

mt

t t

• f

º D

t t t

• q j

º DW º ¢ W

t t

Figure 1. Simulated cosmic-ray anisotropy in equatorial coordinates using the

• M. Ahlers et al (arXiv:1601.07877)
• This can lead to an under- or overestimation of the

isotropic reference level.

Universidad de Guadalajara Centro Universitario de los Valles

The likelihood of observing n cosmic rays is given by the product of Poisson probabilities

relative acceptance relative intensity expected number of events from isotropic background

Maximize the likelihood ratio via null hypothesis in N, A y I maximum values (I⋆,N⋆,A⋆) must follow which can be solved iteratively.

When to stop the iteration?

12 ent distributions ges, .

Iterative maximum likelihood method

Ahlers, BenZvi, Desiati, Díaz-Vélez, Fiorino, Westerhoff (arXiv:1601.07877)

8

• M. Ahlers et al (arXiv:1601.07877)

Universidad de Guadalajara Centro Universitario de los Valles

One-dimensional RA projection of the relative intensity of cosmic rays for adjacent δ bins at -20◦ for HAWC-300 and IC86 data. There is general agreement for large scale

• structures. The two curves correspond to different δ bands but some differences in the

small scale structure might also be attributed to mis-reconstructed events that migrate from nearby δ bins with larger statistics.

9

Universidad de Guadalajara Centro Universitario de los Valles

Significance Map

10

• Relative intensity and significance maps after 20 iterations smoothed over 5deg radius.
• First full-sky combined observation at same energy with two observatories covering

most of the celestial sphere.

• Significance of features in the northern sky is lower than previously published HAWC

results due to decreased statistics from energy cuts.

PRELIMINARY

PRELIMINARY

Relative Intensity

Universidad de Guadalajara Centro Universitario de los Valles

The angular pseudo-power spectrum of the cosmic ray anisotropy for the combined IceCube and HAWC dataset. The gray band represent the power spectra for isotropic sky maps at the 90% confidence level. The structure appears to have a very steep spectrum at low l and a flatter spectrum at l > 3.

11

PRELIMINARY

Angular Power Spectrum

Universidad de Guadalajara Centro Universitario de los Valles

• Iterative method recovers most of the power of large

scale structure in mid-latitude observatories like HAWC.

• No appreciable gain for IceCube.
• The highest angular power for l = 1 is obtained by

combining data from both observatories and using the iterative method.

12

PRELIMINARY

iteration 1 iteration 2 iteration 10

PRELIMINARY PRELIMINARY PRELIMINARY

Angular Power Spectrum

Universidad de Guadalajara Centro Universitario de los Valles

Small-scale Anisotropy

13

PRELIMINARY

PRELIMINARY

Multipole fit

Universidad de Guadalajara Centro Universitario de los Valles

2D Fit

A1 = 1.23e-3 𝜷1 = 35.0 deg. 𝜺0h = 1.005e-3 ± 7.6e-6 𝜺6h = 0.704e-3 ± 7.6e-6

δ0h and δ6h are the dipole components parallel to the equatorial plane and pointing to the direction of the local hour angle 0h (α = 0◦) and 6h (α = 90◦) of the vernal equinox, respectively. The dipole component pointing north δN can not be constrained for ground- based observatories. 14

PRELIMINARY 103 104 105 106 107 energy [GeV] 0.1 1 10 amplitude A1 [10−3] PRELIMINARY

Galactic Center

90 135 180 225 270 315 45 90 phase α1 [degree]

Super-K MACRO IceCube IceTop K-Grande Tibet-ASγ Baksan Milagro EAS-TOP ARGO-YBJ HAWC & IceCube

Summary plot (adopted from M. Ahlers et al. ArXiv:1612.01873) of the reconstructed TeV-PeV dipole amplitude and phase.

Universidad de Guadalajara Centro Universitario de los Valles

Summary

This IceCube-HAWC study is the first (nearly) full-sky cosmic ray arrival direction distribution analysis with combined data from observatories in the North and South at the same primary energy of 10 TeV and is a key to probe into the origin of the CRA observations. Iterative maximum-likelihood reconstruction method simultaneously fits CR anisotropies and detector acceptance. Provides an optimal anisotropy reconstruction and the recovery of the dipole anisotropy for ground-based observatories located in middle latitudes. Ground-based observatories are generally insensitive to cosmic-ray anisotropy variations that are symmetric in RA, i.e. only vary across declination bands (i. e. dipole only observed as a projection onto celestial equator). Nearly full-sky coverage gives better fit of phase and amplitude of horizontal component of the dipole anisotropy. Significant small-scale structure is largely consistent with previous individual measurements. Currently analyzing an additional year of HAWC data which will double the statistics in the northern hemisphere. 


15

Backup

16

Universidad de Guadalajara Centro Universitario de los Valles

IC86 2011-2015

• Data duration:
• 5 continuous years
• 2011/05/13 to 2016/05/20.
• Reconstructed a South Pole with fast but

not very precise algorithms before being transmitted.

• Select long tracks with better angular

reconstruction.

• Reconstructed energy < 32 TeV.

17

HAWC300

• Data duration:
• 269.0 cont. day periods
• (336.36) total days
• 2015/04/21 to 2016/04/19.
• Combine short runs:
• time gaps < 20 min.
• Select high quality reconstructions.
• Eliminate 𝜹-ray candidate events
• Reconstructed energy > 10 TeV.
• log10(Ereco) < 4.5 (32 TeV)
• rlogl* < 15
• ldir_c** > 200 cos(𝛴zenith)
• ndir_c*** > 9 cos(𝛴zenith)

Data Cuts

• log10(Ereco) >= 4.0 (10 TeV)
• nHit* >= 75
• 0 ≤ 𝛴zenith < 1.0 (57.3°)
• CxPE40XnCh** > 40
• PINC*** > 1.6

*Reduced log-likelihood of the track reconstruction fit. **Length of track in direct (on-time) PE hits. ***Number of direct (on-time) PE hits. *Number of PE hits. **Number of channels beyond 40m from the reconstructed core. ***Gamma/Hadron separation (smoothness of shower).

Data Cuts

Universidad de Guadalajara Centro Universitario de los Valles

18

log(E/GeV)

2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 0.000 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040 HAWC IceCube

PRELIMINARY

Top left:The chemical composition as determined by Monte Carlo weighted with Hoerandel polygonato model. Composition differs slightly. This might be attributed to the effect of energy and quality cuts as well as different detector sensitivities to different nuclei. Right: Both CR data sets have a median energy of ~ 10 TeV with little dependence on zenith angle after cuts.

Composition

Universidad de Guadalajara Centro Universitario de los Valles

In the case where you have multiple observations with common pixels, the

• ptimal fit of the relative acceptance 𝒝 (0) and the background rate 𝒪 (0)
• f the null hypothesis becomes

Where s corresponds to the index of the observatory. Then the maximum of the signal hypothesis obeys the implicit relation

Generalization for multiple observatories

N s (0)

τ

= X

i

ws

i nτi ,

As (0)

i

= X

τ

ws

i nτi

. X

κj

ws

jnκj .

I?

a =

X

⌧

n⌧a . X

s

As ?

aN s ? 

, N s ?

⌧

= X

i

ws

i n⌧i

. X

j

As ?

j I? ⌧j ,

As ?

i

= X

⌧

ws

i n⌧i

. X



N s ?

 I? i .

19