Oct 2006 g.modestino1 Experimental results correlating GW detector data and Gamma Ray Bursts...

oct 2006 g.modestino 1

Experimental results correlating GW detector data and Gamma Ray

Bursts

Giuseppina ModestinoLNF

ROG CollaborationINFN – LN Frascati, LN Gran Sasso, Sez. Roma 1, Roma 2 and Genova Universities “La Sapienza” and “Tor Vergata” Rome, L’Aquila, Geneve

CNR –IFN RomaINAF –IFSI Roma CERN - Geneve


Summary• Search for GW Burst Signals

and resonant cryogenic detectors.

• GW and GRB association EXPERIMENTAL IMPLICATIONS.

•Data analysis under two hypotheses•COSMIC RAYS DETECTION analogies and calibration

•Experimental results Previous and in progress


GW Burst Sources

• Binary system coalescence• Supernovae

• ms pulsars

Search for GW Burst Signal

SNR ~ 103 (10kpc/d)2 (10-44 Hz-1/h2) (f /1Hz)

d = distance from earth h = detector spectral densityf = bandwidth of detection


Cryogenic resonant detectors GW detection sensitivity

• ALLEGRO• AURIGA

• EXPLORER•NAUTILUS


SNR ≥ 10 for a galactic event

h = 10-21 ÷ 10-20 Hz -1/2

f = 1 ÷ 50 Hz


Initial operation of the International Gravitational Event CollaborationG.A. Prodi et al. (IGEC Collaboration)Int. Jour. of Modern Physics D 9, 237 (2000)

Study of coincidences between resonant gravitational wave detectorsP. Astone et al. (ROG Collaboration)Class. Quantum Grav. 18, 243–251 (2001)

First search for gravitational wave bursts with a network of detectorsZ.A. Allen et al. (IGEC Collaboration)Phys. Rev. Lett. 85, 5046-5050 (2001)

Search for gravitational wave bursts by the network of resonant detectorsP. Astone et al. (IGEC Collaboration)Class. Quantum Grav. 19, 1367–1375 (2002)

Study of the coincidences between the gravitational wave detectors EXPLORER and NAUTILUS in 2001P. Astone et al. (ROG Collaboration)Class. Quantum Grav. 19, 5449–5463 (2002)

Methods and results of the IGEC search for burst gravitational waves in the years 1997–2000P. Astone et al. (IGEC Collaboration)Phys. Rev. D 68, 022001 2003



Typical theoretical prediction for GW burst amplitude:l/l = h = 3 10-20 1kHz 1Mpc (MGW/10-2 Mo)1/2 (10-3s/GW)1/2

f(kHz) R(Mpc)

0.

1.

2.

[1055

erg/

s]

Gravitational wave luminosity

2 4 6 Time [ms]

Ruffert et al.,AA 311,532 (1996)

associated to a single GRB event: @ 1 kHz

GW = 1ms

10-22 (MGW)1/2

R(Gpc)h ~


GW Burst Detection Strategy

• List of candidate events obtained selecting (with an adaptive threshold) data produced by filter matched to a delta

• Search for coincidences

Data Analysis

Experimental ToolsStatistical excessLocal mass distribution for the directional studies

+

time reference Statistical association

Link with GRBs(GRB=systematic phenomenon)


ASSOCIATION WITH SUPERNOVAE

•Measurements with the resonant Gravitational Wave detector EXPLORER during the GRB 980425 No anomaly in the background of the GW data detector with the sensitivity of h>=10-18

(ROG Coll.) Astron. Astrophys. Suppl. Ser. 138 605-606 1999

•Search for GW associated with the GRB030329 using the LIGO detector.None candidate event detected in the signal time region of 180s [ -120s, + 60s ] around t (LIGO Coll.) Phys.Rev. D 72, 042002 (2005)


Statistical association •Search for Time Correlation Between GRBs nd Data from the Gravitational Wave Antenna EXPLORER.Coincidence technique.No evidence in a time window of + 1 s , at several delays. (ROG Coll.) Astron. Astrophys. Suppl. Ser. 138, 603 –604 (1999)

•Correlation between GRBs AND GWs.Using 120 GRBs, in a 10s time window, an U.L. of 1.5 10-18 was obtained. (AURIGA Group) Phys.Rev. D 63, 082002 (2001)

• Search for Correlation between GRBs detected y BeppoSax and gravitational wave detectors EXPLORER AND NAUTILUSCross-correlation technique-Absence of signal of amplitude h>1.2 10-18 within + 400 s. (ROG Coll. )Phys. Rev. D 66, 102002 (2002) h>6.5 10-19 + 5 s.

• Cumulative analysis of the association between the data of the GW detectors NAUTILUS and EXPLORER and GRBs detected by BATSE and BeppoSAX.Analysis of the data relative to a large number of GRBs detected during the ‘90s, allows to exclude GW burst signal with h> 2.5 10-19

(ROG Coll. )Phys. Rev. D 71, 042001 (2005)

Experimental studies:


Zero-threshold search:better sensitivity than the "event" method for the expected "small" signals.

1)Combine the GW detector data E(t) with the same relative time with respect to the N GRB arrival time t0

E1

E2

E3

En

t(s)ttGRB

2)Applying a cumulative

algorithm: = i N-1/2

For the single data stretches i = <Ei>


EXPLORER is equipped with 3 layers of Plastic Scintillators2 above the cryostat - area 13m2 1 below -area 6 m2)

NAUTILUS is equipped with 7 layersof Streamer tubes 3 above the cryostat - area 36m2/each 4 below -area 16.5 m2/each)

Cosmic Ray detectors

Studing the effects of the showers on the crygenic bars


Unfiltered signal (V2)

The signal after filtering (kelvin)

The effect of the Cosmic Ray Shower on NAUTILUSA true mechanical pulse

The cosmic ray effect on the bar is measured by an offline correlation, driven by the arrival time of the cosmic rays, between the observed multiplicity in the CR detector and the data of the antenna, processed by a filter matched to signals


The effects of th CR on resonant GW detectors

Thermo-acoustic model


kelvin

Time [s]

E ~ 3. 10-4 kelvin

Zero-threshold searchCosmic-ray showers interacting with NAUTILUS

Astone et al. (ROG Coll.)PRD, 84, 14, 2000.

Astone et al. (ROG Coll.)Phy. Lett. B 499 16 (2001)

Astone et al. (ROG Coll.)Phys. Lett. B 540 179 (2002)

2001

In agreement with the thermo-acoustic model

<E(t)>= 3 10-3K ~ 10-4 K

2000


Time delay between the two emissions

The typical amplitude of a GW burst, on Earth, associated to a single GRB, is:

It implies an intrinsic difficulty relating the upper limit on the amplitude,in general, interpreting the measurement results.

10-22 M1/2 R(Gpc)

But there is no clear predictions about t=|tGRB-tGW|

h =R = distance M = amount of GW energy emitted (Mo)

By experiment, it is possible to solve the problem, correlating the data of 2 GW detectorsEssentially, two methodologies are possible:Coincidence analysisCross-correlation technique (advised in nonstationary noise conditions)


Searching for GW signals under variable delay hypothesis

The basic idea of the analysis is the simultaneity of the signal on two GW detectors.

•Apply the usual data selection criteria evaluating the temperature noise in the time interval around the GRB time

•Select and cross-correlate data of the coincident intervals of the two GW detectors.

•Apply the usual cumulative algorithm to the cross-correlation function

Phys. Rev. D 65 022005 (2002)Phys. Rev. D 66 102002 (2002)


EXPLORER(CERN) ON from March to

December T = 2.6 K Duty Cycle = 91% Average sensitivity

h=4.5 10-19

NAUTILUS (LNF) ON from January to

December T = 1.5 K Duty Cycle = 80% Average sensitivity

h=5. 7 10-19

Search for GW signal in a 800 s interval time Beppo-Sax Catalog 2001


.

Cross-correlation analysis

.

The data with Teff < 20 mk are selected.

47 GRB are analyzed

Teff evaluated within a time window of + 400s, centered at tGRB

EXPLORER

NAUTILUS

oct 2006 g.modestino 19(ROG Coll), Phys. Rev. D 66 102002 (2002).

Cross-correlation result EXPLORER x NAUTILUS 2001

R(t’) cross-correlation function (averaged on the 47 GRBs) relative to the energy of the two GW detectors.

-400s < t’ < +400s h <1.2 10-18 95%t’(s)


Searching for simultaneous GW signal.

•1991-1999

•GRB list = BATSE + Beppo-Sax

•GW detectors = EXPLORER +NAUTILUS

•GWB arrival time = GRB Peak Time ± 5 s

•GW data background over 30 minute periods with temperature noise tn< 15 10-3 kelvin(Peak Time ± 15 min)

•Cumulative average and median algorithmROG Coll. PRD 71, 042001 (2005)

Zero-threshold search


GRBs Database

0

100

200

300

400

500

1990 1992 1994 1996 1998 2000 2002 2004

B

BeppoSAX (~1000 )

BATSE (~2700)

Trigger GRBs

ROG Coll. PRD 71, 042001 (2005)


oct 2006 g.modestino 22ROG Coll. PRD 71, 042001 (2005)

Data selection

Effective temperature Te distribution computed in 1150 time intervals around each GRB peak time

N=387 data stretches with Te < 15 mK are selected


E(0) = 6.33 mk


<E> = 6.30 mK

= 0.13 mK

ROG Coll. PRD 71, 042001 (2005)

GW detector energy as a function of the GW-GRB delay

• Combine 387 30 min GW data stretches• For each time delay, the median value Em is extracted

h < 3 10-19


sin4 correspondingly to the 387 selected GRB arrival times and theoretical isotropic distribution The 4 regions of increasing sin4 separated by vertical lines, correspond to the data subsets separately analyzed.

Searching for a correlation with sin4

ROG Coll. PRD 71, 042001 (2005)

~ sin2


4/4

ROG Coll. PRD 71, 042001 (2005)

Looking for a correlation with sin4

SNR of the excess at zero delay of the median value


GW resonant detectorsEXPLORER + NAUTILUS

And

SWIFT events

An example :

Measuring @ t = trigger time+0.5sYear 2005 events with z < 1Interval time for background evaluation :100s

In progress:Data Analysis


Data TakingData Taking during 2005 during 2005


EXPLORER and NAUTILUS -EXPLORER and NAUTILUS - 20052005

NAUTILUS

EXPLORER

10-21 Hz-1/2

~ 50 Hz

@10-20 Hz-1/2

EXPLORER

~ 50 Hz@10-20 Hz-1/2

~ 30 Hz@10-20 Hz-1/2


Output of the cumulative algorithm applied to EX+NAj=1,23 N=23 SWIFT events (11 EX+12 NA)

K

S

trigger time

Ej(t) / N

Red -Sampling time 3.2 ms Black - Integrating on 0.5 s

In progress:


GW resonant detectorsEXPLORER + NAUTILUS

And

Cosmic Rays

Analyze a similar (statistically) sample:

Measuring @ t = CR trigger time+0.5sYear 2005 26 CR events (with multiplicity>2000 part/m2)Interval time for background evaluation :100s

Testing with Cosmic Rays DetectionApply the same procedure:

In progress:


Output of the cumulative analysis applied to EX+NAj=1,26 N=26 EAS events (13 EX+13 NA)

K

S

EAS trigger time

Red -Sampling time 3.2 ms Black - Integrating on 0.5 s

Ej(t) / N



Sensitivity of bar detectors


Sensitivity of the bar detectors

IGEC2



Upper limit for GW burst detection Class. Quantum Grav. 23 S57-S62 (2006)


NAUTILUS

EXPLORER

bar length L = 3mmass = 2300 Kgmaterial Al5056resonance ~1 KHztemperature = 0.1 K - 2 K

ROG - The two cryogenic detectors -

LNF-Frascati

CERN

Oct 2006 g.modestino1 Experimental results correlating GW detector data and Gamma Ray Bursts...

Documents

Transcript of Oct 2006 g.modestino1 Experimental results correlating GW detector data and Gamma Ray Bursts...