1 The Giant Radio Array for Neutrino DetectionExpected sensitivity to cosmic neutrinos: a preliminary estimate

2 A radio telescope dedicated to neutrino detectionToy setup: 90’000 antennas deployed over 220x270km² in Tianshan Giant Radio Array for Neutrino Detection (GRAND) +150 +100 +50 -50 -100 -150 Easting [km] Northing [km] Tianshan mountains 60’000km²

3 Neutrino sensitivity studyEnd-to-end MC simulation: Neutrino trajectory Neutrino interaction in rock (PYTHIA) Tau energy losses (GEANT4) Tau decay (PYTHIA+TAUOLA) Shower development (CONEX) Radio signal generation (EVA) Antenna response (NEC2) Trigger (TREND DAQ) Radio simulation issues at large zenith angles + huge CPU requests [on-going work]. For now simplified hypothesis for shower detection: Antenna fired if: In direct view of shower in a light cone of few degs (=f(E)) Tau decay point at 5km+ Detection if : one cluster of 5+ antennas fired. Shower energy > eV / 1017eV

4 GRAND EAS detection criterium (3): maximum distance to showerExperimental situation ANITA Baloon-borne experiment above the Antarctic. Detection of 16 EAS (14 reflected on the ice surface) with = eV, ~100km from reflexion point. ~100km ANTARCTIC ICE CODALEMA Array of radio antennas on ground Detection of 1017eV showers (with e=85%) ~ 200m away from axis. e(d) =ke0 exp(-d/d0) and e a E

5 GRAND detection criterium (3)Beamed emission within cone of angle ath Neglect signal attenuation within cone (OK if signal strength a Rmax /R with Rmax >10 kms) e(d) =ke0 exp(-d/d0) = ke0 exp(-tanaL/d0) e(a) = ke0 exp(-tana/t) ath : angle above which field is not detectable Efield e / longitudinal distance L E=1017eV Xmax ~ 630g/cm²  L = 6000m tan 𝛼 𝑡ℎ = 𝜏 ln 𝐸 tan 𝛼 17 𝑡ℎ L = 6000m d0 = 400m for q>40° = 0.033 a17 = atan(300/6000) = 3° 300m Distance to shower axis d

6 TREND simulations Simulation chain used for the TREND-50 data (EVA+Conex) [See Valentin’s talk + TREND-50 tomorrow] E = eV simulated showers yields detectable signal q = 80°  shower 30-40km  TREND50 experimental data does not agree so well! TREND50 simulation CONEX+EVA TREND50 experimental data TREND50 simulation CONEX+EVA

7 GRAND EAS detection criterium(1): shower topologyConsider shadowing effect (only antennas in direct view of shower) Discard isolated antennas (dclosest>2km) Request 1+ cluster with 5 antennas at least.

8 GRAND EAS detection criterium (2): minimum distance to showerMinimum shower distance: Shower has to be distant enough to develop and produce enough e+/e- to generate sizeable electromagnetic field. 5km seems reasonable. @ 2000m asl: r = 0.1 g/cm3 atm depth [g/cm²]  0.1xlength [m]

9 TREND EAS detection criterium (4): minimum shower energyEthreshold? CODALEMA: 1017eV. For ~horizontal (fully developped) showers & East-West+North-South measurements, geomagnetic effect should be more efficient. Beamed emission + low attenuation: threshold should not be affected too much by distance to shower. Eth in [3.1016, 1017] eV

10 Simulation results ~ Horizontal trajectories.Mountains are sizable tragets. Many extended tracks ( = 190 antennas). Angular resolution: < 0.1° (assuming Dt~1ns) eV eV 3.1019eV 3.1020eV Downward Going (mountains) Upward (Earth) PRELIMINARY

11 GRAND neutrino sensitivityPRELIMINARY GRAND : 90% CL limit assuming 0 candidates in 3 years threshold = eV F = F0E-2 spectrum Tens of GZK n /year expected! th = 1017eV th = eV GRAND could reach ~5x better sensitivity than Antartica projects. To be checked/optimized with full MC.

12 Signal characteristicsTrigg’d antennas far away from tau decay point: = 50km. Large number of antennas triggered: Angular resolution~0.1° if dt~1ns Mult≥8 affects detection marginally (-15% in sensitivity) Antenna step = 800m OK (-25% compared to 400m step). Median mult = 20 antennas Nb of antennas per trigd cluster

13 Signal characteristicsTopology as a critical parameter for detection: there is room for layout optimisation! Altitude (m) Nb of triggers/antenna (log scale)

14 Background rejection Atmospheric neutrinos HE muonsNegligeable at high energy HE muons Back of the enveloppe calculation based on arXiv:hep-ph/ : decays/year over full array Standard cosmic ray EAS Cut 1° below horizon (mountains ) 1°  5s for 0.2° angular resolution suppression factor Discrimination on young (neutrinos) vs old (CRs) showers (?) Affects marginaly detection efficiency: <10% ICECUBE astro-ph: 0.2° angular resolution at worst for st = 1ns 5 antennas triggering

15 Terrestrial backgroundAll radio projects (AERA, TREND…) exhibit large trigger rates, even in remote areas. Terrestrial origin: HV, train, planes, thunderstorms… Trig rate is orders of magnitudes larger than expected neutrino rate. Rejecting this background is a (the?) major challenge for the GRAND project.

16

17 t nt Neutrino detectionExtensive air shower t nt Radio detection Eth ~1017 eV Earth + mountains as target for neutrino interaction Fargion et al, astro-ph/ , Bertou et al., astro-ph/ Radio detection of subsequent EAS (good at large zenith angles)

18 GRAND hot spots Hot spots: zones with optimal topology for neutrino ddetection. Event rate ~5x larger than average. Detected events intensity map

19 GRAND neutrino sensitivityThreshold = 1017 eV

20 GEANT4: t Energy Loss Typical interaction length from GEANT4, with all processes (Ionisation, Bremsstrahlung, pairs, photonuclear) Photonuclear loss is dominant at UHE and typical interaction length  b Theoretical expectation for b parameter in photonuclear processes (following Dutta et al.)

21

22 TREND detection criterium (3)ath17 ? CODALEMA showers: dmax ~ 300m (CODALEMA) Xmax ~ 1017eV and =30° L~6000m ath17 = atan(dmax/L) ~ 3° t? d0 in [100,400m for L~6000m ] (CODALEMA) t = d0/L in [0.017, 0.067] q=30°

23 Terrestrial backgroundGRAND bckgd event rate estimation: TREND50: ~30kEvents/day/km² (~ valid coincs in 317 DAQ days) TREND-50 event multiplicity events TREND50 → GRAND: Size x40000 Antenna density /25 7500 events GRAND: <15 events/day/km² (safe estimate) 3 108 evts/year over full array Expected n event rate: 0-100 events/year Target rejection factor: 109

24 How far can we see EAS radio signals?Radio detection ultimate threshold: 0.5µV/m/MHz (Galactic noise) Lateral distribution: e=e0exp(-d/d0) with d0~200m (or more) e0 scales linearly with E after correction for geomagn. angle (data & simu) and 1017eV  0-2 µV/m/MHz (let’s take 1 on average). 𝜖 µ𝑉 𝑚 −1 𝑀𝐻𝑧 −1 = 𝐸 𝑒𝑉 exp − 𝑑 𝑚 > 0.5 necessary for detection CODALEMA data, V. Marin PhD (2013 SUBATECH)

25 How far can we see EAS radio signals?Radio detection ultimate threshold: 0.5µV/m/MHz (Galactic noise) Lateral distribution: e=e0exp(-d/d0) with d0~200m (or more) e0 scales linearly with E after correction for geomagn. angle (data & simu) and 1017eV  0-2 µV/m/MHz (let’s take 1 on average). 𝜖 µ𝑉 𝑚 −1 𝑀𝐻𝑧 −1 = 𝐸 𝑒𝑉 exp − 𝑑 𝑚 > 0.5 necessary for detection CODALEMA data, V. Marin PhD (2013 SUBATECH)

26 How far can we see EAS radio signals?a = 750m OK for E>1019 eV (1000m for E>1019,5 eV). Conservative estimate!!! (6+ antennas, q=0° & d0=200m) 4 x denser than present AUGER array… but loads more cheaper & easier! Inversion gives max distance allowed for detection: To get 6+ antennas with this condition, we need an array step length a such that : 6a² < pd²  𝑑<200 log 2𝐸 𝑒𝑉 𝑚 a d a< 200 𝜋 √6 log 2𝐸 𝑒𝑉 𝑚

27 TUNKA Rex (63000km²)