Acoustic detection of high energy
neutrinos in sea water: status and
prospects
or: the road to KM3NeT
Robert Lahmann
ARENA 2016
Groningen, 8-June-2016
Outline
● 
● 
● 
● 
Introduction: sound in water and acoustic neutrino detection
The “first generation” of acoustic neutrino test setups
Results and lessons learned
Conclusions and outlook
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
2
Sound in water
“Of all the forms of radiation known, sound travels through the sea the
best”
(R. Urick, Principles of Underwater Sound, 3rd edition, 1967)
Used by marine animals and humans
for communication and positioning
Speed of sound in water investigated
(at least) since 1826 (from title page of
“Physics Today”, Oct. 2004, experiment
in Lake Geneva)
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
3
Absorption Length [km]
Sound absorption length in water
4
10
3
10
Mediterranean
2000m
100m
2
10
1
10
+
NaCl ⇔ Na Cl
~5km
−
fresh water
0
10
−1
10
MgSO4 ⇔ Mg 2+ SO42−
−2
10
0
10
1
10
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
2
10
3
10
Frequency [kHz]
4
Acoustic signals of neutrino interactions in water I
Thermo-acoustic effect: (Askariyan 1979)
energy deposition ð local heating (~µK) ð expansion ð pressure signal
Wave equation for the pressure p for deposition of an energy density ε :
2
2
1
∂
p
α
∂
ε
2
∇ p− 2 2 =−
c ∂t
C p ∂t 2
α
= Volume expansion coefficient
C P = specific heat capacity (at constant pressure)
c
= speed of sound in water (ca. 1500 m/s)
Solution (analytical/numerical) with assumption of an instantaneous energy deposition
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
5
Acoustic signals of neutrino interactions in water II
0.12
~1o
ν
0.1
Bipolar
Pressure
Signal
(BIP)
0.08
Pressure [Pa]
[Pa]
Pressure
Hadronic cascade:
~10m length, few cm radius
(simulations by ACoRNE Coll.)
0.06
0.04
0.02
0
-0.02
-0.04
-0.06
-0.08
1011GeV @ 1000m
-0.04
-0.02
0
0.02
Time [ms]
Time
[ms]
0.04
Pressure field:
Characteristic “pancake” pattern
Long attenuation length (~5 km @ 10 kHz)
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
6
The bipolar pulse
ν
σρ
Analytical calculation of a
signal in the far field for a
Gaussian energy density
2 pmax
E0 = ∫ ε dV
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
E0
pmax ∝ γ G 2
σρ
c 2α
γG =
CP
c = c(temp, pressure, salinity)
α = Volume expansion coefficient
C P = specific heat capacity (at constant pressure)
c = speed of sound in water (ca. 1500 m/s)
7
0
−500
Depth [m]
Depth [m]
Temperature profile
winter
summer
−1000
0
−1000
−2000
−1500
−3000
−2000
−4000
−2500
12
14
16
18
20
22
24
Temperature [C]
−5000
0
5
10
15
20
25
30
Temperature [C]
(a)
Tropical(b)
Ocean
Mediterranean Sea
24o30’N and 72o30’W
Figure 4.7: Temperature as a function of the depth below the sea surface, for different seasons
at two different sites: (a) Measurement from the ANTARES site in August 2007 [170] (solid
blue line)
and in March 2010 [171] (dashed red line); (b) Temperature profile in the tropical
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
8
◦
◦
ocean [172], 24 30’N and 72 30’W, for the average of the three summer months (solid blue
line) and the three winter months (dashed red line).
0
Depth [m]
Depth [m]
Speed of sound vs. depth
−500
0
−1000
−1000
−2000
−1500
−3000
winter
summer
−2000
−2500
1500
1510
1520
−4000
1530
1540
1550
−5000
1490
1500
1510
1520
Speed of sound [m/s]
(a)
1530
1540
1550
Speed of sound [m/s]
(b)
Tropical Ocean
Mediterranean Sea
o30’N and 72o30’W
24the
Figure 4.8: Speed of sound as a function of the depth below
sea surface, for the seasons
and sites as in Fig. 4.7 derived using the parametrisations of water properties from [169]: (a)
Profile
of theGroningen
speed- 8-June-2016
of sound
at the
ANTARES site in August 2007 [170] (solid blue line)
ARENA2016
- Robert
Lahmann
9
Grüneisen parameter
0
Depth [m]
−1000
−2000
−3000
−4000
−5000
−0.05
tropical ocean
Lake Baikal
AMADEUS site
0
0.05
0.1
Gr üneisen parameter
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
0.15
0.2
10
Acoustic detection test setups
First generation acoustic test setups for feasibility studies (background),
developing techniques/algorithms
Two “philosophies”:
●  “We can get access to an acoustic array; why not use it for some
tests for acoustic particle detection?”
●  “We have a neutrino telescope infrastructure; why not install some
acoustic sensors to test acoustic particle detection?”
Technology:
Hydrophones (in water) and glaciophones (in ice) using piezo ceramics
Array size:
O(10) sensors
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
11
Test Setups in ice and water
ACORNE (M) AMADEUS
(ANTARES)
KM3NeT-Italia/OnDE
Baikal
SAUND (M)
green:
currently in operation
(M):
military array
salt water
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
SPATS (IceCube)
12
SAUND
II
S
SAUND
AUTEC
SAUND and
AUTEC and
49 Uni-directional Hydrophone readout
20 x 50 km array
15
12
SAUND II :
49 hydrophones
on
20km x 50km
array
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
13
OnDE and KM3NeT-Italia
•  Test Site at 2000 m depth, 25 km
offshore Catania
•  Operation of test setup OnDE
(4 hydrophones) from 2005 -2006
Cable from shore
H2
hydrophones
H1
H3
electronics
housing
Housing
H4
connectors
Height from seabed :
H1, H2, H4: ~ 2.6 m
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
H3: ~ 3.2 m
KM3Net-Italia activities
covered by F. Simeone
14
The Rona Array (ACoRNE Collaboration)
Off the Isle of Skye, 8 sensors
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
15
AMADEUS – ANTARES
ANTARES
site
Operational 2007-15
36 acoustic sensors on
6 stories
Local clusters for
direction reconstruction
Depth 2300 – 2100 m
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
16
SPATS, ARENA 2012
Limits on UHE neutrino flux
SPATS
ACORNE
SAUND2
R. Abbasi et al.,arXiv:astro-ph/1103.1216
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
17
OnDE ambient noise (ARENA 2008)
Main source: Surface agitation and precipitation Average noise 2005-2006
Sea State 2
Sea State 0
astro/ph 0804.2913
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
f2
Ap ( f1 , f 2 ) = ⎡ ∫ PSD ⋅ ( f )df ⎤
⎢⎣ f1
⎥⎦
1
2
The average noise in the [20:43] kHz
band is 5.4 ± 2.2stat ± 0.3syst mPa
18
AMADEUS ambient noise
Simulation
60000
50000
Entries
entries
1
Frequency of Occurence
Cumulative Distribution
0.9
0.8
0.7
40000
0.6
30000
0.5
0.4
20000
0.3
0.2
10000
0.1
0
Data of 2008-2010
0
1
2
3
4
5
6
7
σNoise/< σNoise >
8
9
10
0
<σnoise> is about 10 mPa (10-50 kHz) and 95% of the time below 2<σnoise> ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
19
Probalilty
Measurement
Transient background
•  Sources: Very diverse;
Shipping traffic, marine mammals, …
ð Mostly originating from near surface
•  Suppression:
•  signal classification
•  Project reconstructed signals to surface,
perform clustering
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
20
Cluster analysis of moving sound emitting objects
y [m]
Oct.
2010
Nov.
2009
x [m]
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
21
1000
500
0
-500
-1000
-1500
-2000
-2500
All clustered
events
All reconstructed
events
0.3 Hz
0.001
0.0001
1e-05
1e-06
1e-07
1e-08
0
2000
4000
6000
Distance r [m]
8000
10000
After signal classification
and cluster analysis
0.002 Hz
0.001
0.0001
1e-05
1e-06
1e-07
1e-08
0
2000
4000
6000
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
Distance r [m]
8000
Event density [m-3]
1000
500
0
-500
-1000
-1500
-2000
-2500
Event density [m-3]
Depth z [m]
Depth z [m]
Spatial distribution of transient background
10000
22
First generation setups: lessons learned
●  Ambient background:
Background low and stable, reduction of SNR for signal detection
crucial
●  Transient noise in the Mediterranean:
High level of background (mainly dolphins);
High level of reduction already achieved with AMADEUS, recognition
of “acoustic pancake” crucial (talk by Dominik Kiessling)
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
23
wate
n
e
p
o
r
The road ahead
●  Second generation acoustic neutrino detection test setup (in Sea
Water):
“KM3NeT will have a huge array of acoustic sensors for position
calibration; why not use it to implement what we have learned with
the first generation setups?”
(see following talk by Francesco Simeone)
●  Third generation acoustic neutrino detector:
KM3NeT can be extended with acoustic sensors using new and
innovative technology (see talk by Ernst-Jan Buis).
Results will be presented at ARENA 20XX
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
24
Conclusions and outlook
●  “First generation” acoustic arrays have been used to investigate
neutrino detection methods and provide input for simulations
●  Working on methods to reduce SNR, recognize “pancake”
●  The future of acoustics in sea water is KM3NeT
ARENA2016 Groningen - 8-June-2016 - Robert Lahmann
25
Thank you for your attention!