Indirect dark matter searches with neutrinos from the Galactic Centre region with the ANTARES and KM3NeT telescopes

(1)

Indirect dark matter searches with neutrinos from the Galactic Centre region with the ANTARES and KM3NeT telescopes

Sara Rebecca Gozzini

on behalf of the ANTARES and KM3NeT Collaborations Instituto de F´ısica Corpuscular (IFIC), University of Valencia and CSIC

May 17, 2021

(2)

Theory (1) - WIMPs

Hints of the existence of dark matter are, as of today, only of macroscopic nature: based on cosmology and gravitation.

It comes natural to theorise a dark matterparticlecandidate. For instance WIMPs.

WIMP miracle: interaction strength is of the same size as the known weak interaction.

Universality: despite numerous models with differences in the details of the interaction. → It is possible to predict fluxes of SM products from WIMPs decay or pair-annihilation.

J

EARTH

astrophysics measurement n dN / dE

particle physics

(3)

Theory (2) - Process

The process that we would like to observe is a pair annihilation of WIMPs producing final fluxes of high-energy ν (directly or through an intermediate channel, still inside the source)

WIMP

SM

Measurement = number of outcoming events →translates into number of processes.

The probability for oneprocess to happen is ∝velocity of projectile ×σ.

Generally: rate outcoming particles∝velocity incident particle×σ ×number of targets WIMPs are cold, non-relativistic →vc, we cannot extract c, we only know a velocity distribution ⇒ measurement (limit) ofvelocity-averaged cross-section hσvi.

(4)

Theory (3) - Which measurement?

We measure (limit) the v-averaged cross-section through the numberof ν-induced events

n u m b e r o f e v e n t s

WIMP

__

n t = 1

2hσvi

Z M_WIMP 0

dN dEdE 1

4πJ 1

M_WIMP² A(M_WIMP)

The detector acceptance is described through the effective area (geometry, efficiency, analysis cuts) modulated by the energy distribution of the single WIMP WIMP collision.

: PPPC4 [hep-ph: 1012.4515]provides the energy distribution per collision ^dN_dE

: We sit inside the source. The J-factor (here obtained using[https://clumpy.gitlab.io/CLUMPY]) provides number of WIMPs in a cone around the source identified by our detector’s FoV.

We can test different dark matter halo profiles: NFW, Burkert, MacMillan, Einasto.

(5)

Instrument (1) ANTARES

In continuous, smooth operation for more than 13 years with very little off-duty time.

Layout: 40 km offshore Toulon, 12 lines, 885 PMTs, 2500 m depth. Designed forν astronomy, pointing accuracy from ∼1^◦ to 0.3^◦ for the energy range of DM analyses [50 GeV - 100 TeV].

(6)

Instrument (2) The multi-site KM3NeT infrastructure

ORCA

ARCA

In modular construction!

ORCA: 1 dense block (8 Mton, 115 lines) for measuring atmospheric ν properties with high statistics. As of today: 6 lines, some in operation for more than 1 year ARCA: 2 very-large volume blocks (1 Gton, 230 lines) to catch astrophysical fluxes. As of today: 6 lines

GC

Good visibility of the Southern sky for upgoing events from Mediterranean Sea, including best dark matter source: the centre of the Milky Way - close source→ high J-factor

Sea water has good optical properties for excellent pointing accuracy (low scattering)

(7)

ANTARES latest data set

January 2007 to February 2020 (3845 days). ANTARES sees two topologies of events:

track-like, coming from CC interaction of ν_µ

shower-like, coming from NC interaction ofν or CC interaction of ν_e,ν_τ

All events are selected as upgoing (through the Earth, good atmospheric muon suppression).

Events are classified as track- or shower candidates through geometry parameters together with the likelihood of their reconstruction fit.

11174 events reconstructed as tracks: good angular resolution

225 events reconstructed asshowers: vertex confinement, spherical topology

Extra selection based on energy proxy N_HIT is applied to exploit the spectral shape of channels like WIMP WIMP→νν where events pile up at high energy.

(8)

Analysis logics

The search for a signal of dark-matter induced neutrinos is conducted with a hypothesis test:

H0 only background of atmospheric neutrinos is contained in the sample H₁ signal ofn events from the GC region is due to dark matter annihilations

background distributions are obtained from blind data

signal is reproduced with simulated data, space-distributed as the DM halo profile and energy-weighted according to 5 channels

1 WIMP WIMP→τ⁺τ⁻

2 WIMP WIMP→µ⁺µ⁻

3 WIMP WIMP→W⁺W⁻

4 WIMP WIMP→bb

5 WIMP WIMP→νν

(9)

Dark matter signal at ANTARES

The signal is reproduced weighting simulated events with PPPC4 tracks×^dN_dE|final−state−νµ×oscillations

showers×^h^dN_dE|final−state−νe/ντ

i×oscillations

e e

GC

detector

102 10³ Eν [ GeV ]10⁴ 5

− 10 4 10−

3

− 10 2

− 10 1

− 10 1 ]-1 [ GeVν/dEνdN

pre oscillations all flavour νµ final state

νe final state

ντ final state channel - W + from W

102 10³ Eν [ GeV ]10⁴

5 10−

4

− 10 3 10−

2

− 10 1

− 10-1 ] [ GeV/dEdNνν

pre oscillations all flavour νµ final state νe final state ντ final state channel τ- τ+ from

102 10³ Eν [ GeV ]10⁴ 8

− 10 7 10−

6

− 10 5

− 10 4 10−

3

− 10 2 10−

1 10− 1 ]-1 [ GeVν/dEνdN

pre oscillations all flavour µ final state ν

νe final state τ final state ν

channel b from b

102 10³ Eν [ GeV ]10⁴

5

− 10 4

− 10 3 10−

2 10− ]-1 [ GeVν/dEνdN

pre oscillations all flavour νµ final state νe final state ντ final state channel µ- µ+ from

102 10³ Eν [ GeV ]10⁴

7

− 10 6

− 10 5

− 10 4 10−

3 10−

2 10−

1

− 10 -1 ] [ GeV/dEdNνν1

pre oscillations all flavour µ final state ν

νe final state τ final state ν channel νµ νµ from

source

102 10³ Eν [ GeV ]104 5

− 10 4 10−

3

− 10 2

− 10 1

− 10 1 ]-1 [ GeVν/dEνdN

post oscillations all flavour

νµ final state νe final state ντ final state

channel - W + from W

102 10³ Eν [ GeV ]104

5 10−

4

− 10 3 10−

2

− 10 1

µ final state ν

νe final state τ final state ν

channel τ- τ+ from

102 10³ Eν [ GeV ]104 8

− 10 7 10−

6

− 10 5

− 10 4 10−

3

− 10 2 10−

1 10− 1 ]-1 [ GeVν/dEνdN

channel b from b

102 10³ Eν [ GeV ]104

5

− 10 4 10−

3

− 10 2 10−

1

channel µ- µ+ from

102 10³ Eν [ GeV ]104

5 10−

4

− 10 3 10−

2

− 10 1 10− ]-1 [ GeVν/dEνdN

νµ final state

νe final state

ντ final state

channel νµ νµ from

detector

(10)

Detector acceptance to dark matter events

Cuts are optimised for best sensitivity. Folding in the corresponding spectra, this results in the acceptance below. The ratio of tracks (showers) to total is later used to simulate the correct fraction for each topology, mass and channel considered.

102 10³ 10⁴ MDM / GeV 10⁵

8 10−

7 10−

6 10−

5 10−

4 10−

−3 10

2 10−

1 10− 2Integrated acceptance / m

Graph

tracks τ- τ+

showers τ- τ+

tracks νµ νµ

showers νµ νµ

tracks b b

showers b b

A= Z

A_EFF(E)dN dEdE

A N T A R E S P R E L I M I N A R Y

(11)

Analysis method for hypothesis test

Unbinned maximum likelihood is used to obtain the most likely number of signal events n^∗_s L=

nTOT

Y

i

[n_s ·P_s(angle,N_HIT, β) +n_b·P_b(angle,N_HIT, β)]

: Signal tracks BG tracks : Signal showers BG showers

TS = log

L(n^∗_s) L_BG

(12)

All-flavour cross-section sensitivities

102 10³ 10⁴ 10⁵

/ GeV MWIMP 25

10− 24

10−

−23

10

−22

10

21

10−

−20

10

-1 s3 v > / cmσ<

Ann. channels µ-

µ+

τ-

τ+

b b -

+ W W

(all fl.) ν ν

Solid: sensitivities 2007-20 all flavour - dashed: limits 2007-17 [Phys. Lett. B 805, 2020]

A N T A R E S P R E L I M I N A R Y

n^tr+sh=

n^tr_ν_e +n^tr_ν_µ+n^tr_ν_τ +n_ν^sh_e +n^sh_ν_µ+n^sh_ν_τ = A^tr_ν_e ·Φ^tr_ν_e +A^tr_ν_µ·Φ^tr_ν_µ+A^tr_ν_τ ·Φ^tr_ν_τ+ A^sh_ν_e ·Φ^sh_ν_e +A^sh_ν_µ·Φ^sh_ν_µ+A^sh_ν_τ ·Φ^sh_ν_τ =^?

Φ_1f ^hA^tr_ν_e +A^tr_ν_µ+A^tr_ν_τ +A^sh_ν_e +A^sh_ν_µ+A^sh_ν_τⁱ= Φ_1f ·A^tr+sh_tot

(13)

Follow-up: KM3NeT

The potential of KM3NeT looking for dark matter was estimated for the Sun (ORCA) where low energies are favoured, and for the Galactic Centre (ARCA) that gains from high energies.

ANTARES 11 years NFW KM3NeT ARCA 230 lines 1 year NFW HESS 10 years GC survey Einasto VERITAS Dwarf Spheroidals NFW Fermi+MAGIC Dwarf Spheroidals NFW IceCube IC86 WIMP GC NFW

1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

10^-27 10^-26 10^-25 10^-24 10^-23 10^-22 10^-21

LogM_WIMP[GeV/c²]

〈σv〉[cm3s-1]

Galactic Centre Sun, spin-dependent Sun, spin-independent

ANTARESandARCA ORCA ORCA

As of today, 6 lines of ORCA are recording data (some for ∼1 year), and 6 lines of ARCA are in operation. Search for DM in first KM3NeT data begins.

(14)

Summary of results

ANTARES has extended their analysis of dark matter from the Galactic Centre using all-flavour data and data set extended from 2007 - Dec. 17 → 2007 - Feb. 2020.

The first KM3NeT data is soon ready to be analysed. 6 detection lines both for ORCA and ARCA will serve as a first testbench of dark-matter sensitivities

Next future

New technique (machine-based pattern recognition algorithms, such as anomaly detection) are being studied to disentangle DM signals

Joint dark matter search with combined ν and γ is being developed