gravitational waves: tool for astronomy

Gravitational Waves:Tool for Astronomy

Bernard F SchutzMax Planck Institute for Gravitational

Physics(Albert Einstein Institute)

Potsdamand

Cardiff University, Cardiff, UK

http://www.aei.mpg.de

http://www.aei-potsdam.mpg.de/

http://www.aei-potsdam.mpg.de/

Bernard F Schutz Albert Einstein Institute

22 January 2008

Gravitational Waves: Tool for Astronomy

2

Gravitational Wave Messengers• Ripples in space-time, traveling at the speed of light• Almost orthogonal to EM waves as messengers:

– Source is mass-energy: universal, cumulative– Coupling very weak: universe transparent to GWs

• Expectations from EM astronomy not precise– Known GW sources not well-modelled or surveyed– ‘Dark matter’ might contain sources that never radiate EM, such

as cosmic strings, BHs

• GW detectors rapidly improving sensitivity (range): – 60 Mpc for merging black holes now– 2x larger 18 months from now– 10x 5 years from now– entire Universe with launch of LISA in ten years.

• Detection more like listening than seeing

Why search for GWs?• Test and verify general relativity• Learn more about known sources• Explore the dark part of the universe

• Because they are there!!

Why search for GWs?• Test and verify general relativity• Learn more about known sources• Explore the dark part of the universe

• Because they are there!!


22 January 2008


3


22 January 2008


4

Tidal gravitational forces

• By the equivalence principle, the gravitational effect of a distant source can only be felt through its tidal forces – inhomogeneous part of gravity.

• Gravitational waves are traveling, time-dependent tidal forces.

• Tidal forces scale with size, typically produce elliptical deformations.


22 January 2008


5

Gravitational Waves


22 January 2008


6

GW physics across the spectrum

810low

high

f

f

A chirping system is a GW standard candle: its

distance can be inferred from GW obs.

3/2

1~

2

1RGMGf


22 January 2008


7

Testing Gravitational Wave Theory• It is now 15 years since Russell Hulse and Joe

Taylor were awarded the Nobel Prize for the discovery and scientific exploitation of PSR1913+16

• This system was the first good test that GR is reliable for most source calculations.

• Now there are about 6 similar systems known, and the spectacular “double pulsar” PSR J0737-3039 is already overtaking 1913 in precision.

• Since the GW frequency of this system is lower than the LISA waveband, any corrections to GR due to “massive gravitons” or similar effects can be neglected for current observational programs.

Taylor

Hulse


22 January 2008


8

Ground-Based Interferometer Network

LSCLSC-V

40 Hz < f < 1 kHz

• Detection confidence• Source polarization• Sky location (FOV ~ 4π)• Duty cycle• Waveform extraction – VLBI with λ ~ 3000 km!


22 January 2008


9

LIGO

• Locations: Hanford WA, Livingston, LA

• Partners: Caltech, MIT (NSF facility)

• Length: 4km, 2km at Hanford; 4 km at Livingston

• Data analysis, detector development done by LIGO Scientific Collaboration (LSC)


22 January 2008


10

VIRGO


22 January 2008


11

GEO600


22 January 2008


12

GEO Mirror Suspension


22 January 2008


13

LISA – Shared Mission of ESA & NASA

• ESA & NASA have exchanged letters of agreement. ESA/ESTEC and NASA/GSFC jointly manage mission.

• Launch 2018 into solar orbit, observing 2020+.

• Mission duration up to 10 yrs.

• LISA Pathfinder will demonstrate the key measurement system: picometer accuracy, disturbance-free at L1 (ESA: 2010)

• Joint 20-strong LIST: LISA International Science Team


22 January 2008


14

LISA in Orbit


22 January 2008


15

Progression of GW Astronomy

• Expectations about detections based on estimates of source strengths and their populations, both uncertain.

• Because data analysis uses source models, range depends on source. Main limitation: detector noise.

• Roughly, if broadband noise h > 10-21, strong sources are very rare (1/30 years). This was LIGO pre-2006.

• LIGO 2006-8 (S5) observing at h ~ 10-21. Analysis on-going. Detections possible but would be surprising.

• Advanced LIGO (2013) will observe at h ~ 10-22. Detections almost certain. Factor of 2 enhancement coming first, in 2009.

• LISA (2018) will have very high sensitivity in mHz band, detections assured, strong tests of GR, precise measurements in cosmology. Limitation: confusion noise


22 January 2008


16

What are we looking for from the ground?

In S5:• Unexpected bursts (hidden supernovae?)• Triggered events: GRBs that might represent mergers of neutron stars, out to 20 Mpc• Black-hole mergers out to 60 Mpc: rate might be sufficient for one every few years• Spinning neutron stars in the Milky Way, whose irregular crusts could radiate GWs• Random background from the Big Bang weaker than nucleosynthesis bound (Ωgw ~ 10-5)

In 5 years:• Neutron star mergers out to 400 Mpc• Black-hole mergers out to 2 Gpc (z~0.5)• Intermediate-mass BH mergers (1000 M)• GWs from LMXBs• Random background Ωgw ~ 10-9.• Exotica: cosmic strings, unpredicted sources


22 January 2008


17

How do we analyse data?

• Each source has different search method, so LSC and VIRGO searches run by 4 teams:– Burst searches– Inspiral searches– Pulsar searches– Stochastic searches

• Compute-intensive searches. Large teraflop clusters at several institutions. Einstein@Home is our most powerful tool, doing searches for unknown pulsars.


22 January 2008


18

Compact binary signals in S5• Twelve months of data analyzed - no signals seen• For 1.4-1.4 Mo binaries, ~ 200 MWEGs in range• For 5-5 Mo binaries, ~ 1000 MWEGs in range• Inspiral horizon for equal mass binaries vs. total mass

Peak- 130 Mpcat total mass ~ 25M

sun

•

(horizon=range at peak of antenna pattern; ~2.3 x antenna pattern average)

Triggered Searches for GW Bursts

GRB 070201

No GWs seen: if the event was a

merger, it was much further away

than M31.


22 January 2008


Joint 95% upper limits for 97 pulsars using ~10 months of the LIGO S5 run. Results are overlaid on the estimated median

sensitivity of this search.

Search for known pulsars- preliminary

For 32 of the pulsars we give the expected sensitivity upper limit (red stars) due to uncertainties in the pulsar parameters .

Pulsar timings provided by the Jodrell Bank pulsar

group

Lowest GW strain upper limit:PSR J1802-2124(fgw = 158.1 Hz, r = 3.3 kpc)h0 < 4.9×10-26

Lowest ellipticity upper limit:PSR J2124-3358(fgw = 405.6 Hz, r = 0.25 kpc) < 1.1×10-7

Preliminary

Now about 3 x below

Crab spindown limit.


22 January 2008


21

Pulsar Data Analysis• Wide-area searches for

unknown GW pulsars are computationally very demanding: must use several months continuous data, demodulated for each 1-arc-second patch on the sky.

• LSC and Virgo have developed powerful and efficient hierarchical search methods based on pattern-finding with the Hough Transform, originally devised for analysing bubble-chamber photographs in HEP. (See arXiv, or e.g. Phys. Rev. D72 (2005) 102004 )

• These run on clusters and also on Einstein@Home. E@H delivers 70 Tflops continuous computing.


22 January 2008


22

Limits on isotropic stochastic GW signal

• LIGO S4: ΩGW < 6.5x10-5 [new upper limit;

accepted for publication in ApJ] bandwidth: 51-150 Hz;

• Initial LIGO, 1 yr data, expected sensitivity ~

4x10-6

• Upper limit from Big Bang nucleosynthesis 10-5; interesting scientific territory

• Advanced LIGO, 1 yr data, expected sensitivity ~1x10-9

GWs

neutrinos

photons

nowCosmic strings (?) ~10-8

Inflation prediction ~10-14


22 January 2008


23

Future Plans – Ground Based• LIGO is now shut down for upgrade to Enhanced LIGO, will

resume observing mid-2009 with ~ 2 × increased range. This may be enough for, eg, binary BH systems.

• VIRGO hopes to match Enhanced LIGO in 2009.• GEO and H1 will observe during LIGO upgrade, then in

2009 GEO will begin upgrade to GEO-HF, with enhanced high-frequency sensitivity (1-2 kHz), aiming at neutron star normal mode frequencies.

• In 2011 LIGO and VIRGO will begin upgrades to Advanced detectors, online again by end 2013. These have essentially guaranteed sources in NS-NS coalescences.

• Design study for 3rd-generation Einstein Telescope has been funded in Europe by EU/FP7. This is where GW astronomy will become a major input to a broad range of astrophysics.


22 January 2008


24

Vibration control limit

Shot noise limit

Arm-length limit

LISA Sensitivity


22 January 2008


25

What science will we do with LISA? Mergers of supermassive black holes:

• Detect all mergers in 104 M < M < 108 M

• Measure distances, distribution of masses, evolution of population with time (with

z)• Measure Hubble expansion rate, acceleration, evolution of dark energy: all with no need to calibrate distance ladder• Determine whether SMBHs formed before, with, or after galaxies, and whether they grew by accretion or by BH merger• Compare mergers with numerical simulations, test strong-field GR, Hawking area theorem

Other sources:• Discover all compact binaries in Galaxy with periods < 1 hr• Measure mass function of white dwarfs• Using EMRIs, test GR: are all BHs Kerr? • Use EMRIs to probe stellar distribution near SMBHs, evolution of central parts of galaxies• Random background Ωgw ~ 10-10.• Exotica: cosmic strings, unpredicted sources: searches at high signal to noise!


22 January 2008


26

Massive Black Holes Exist!

Genzel, et al, MPE


22 January 2008


27

Massive Black Holes Merge• Known masses from

106 to 109 M. Smaller masses possible.

• Galaxy mergers should produce BH mergers. Rate uncertain, but several per year in Universe possible.

• Proto-galaxy mergers may create thousands per year of smaller (104 M) BH mergers.

(Chandra Observatory)

NGC 6240


22 January 2008


28

BH Merger Simulations

• Numerical relativity simulations are now producing accurate solutions of Einstein’s equations for a large variety of in-spiraling systems of black holes with spin.

• Key physics question is how large is the “kick” that the remnant receives from the asymmetrically radiated GWs. Record is 2000 km/s, typical seems to be 100-400 km/s. Depends very sensitively on mass ratio and spin directions.

• Speculation that maximum final J limited to ~ 0.7 M2. W Benger/AEI/ZIB


22 January 2008


29

Data Analysis Challenges for LISA• Huge parameter space for capture orbits• Strong interaction between searches for different

sources:– SMBH signals distort noise, must be

removed well– Binary confusion noise “edge” must be

removed, improves with time– Capture orbit confusion may have to be

done simultaneously• Careful planning, plus prior research and simulation

essential!• Coordination with astronomical community equally

important.


22 January 2008


30

LISA’s Future• Joint ESA-NASA mission. In ESA’s schedule LISA starts at

the end of LPF, 2012, and launches 2018.• LISA Community (LISC) website has information and

latest news:http://www.lisa-science.org/

• NASA recently requested an review of its priorities in its Beyond Einstein Program (including LISA) from a committee of National Academy. Recent BEPAC report (Preliminary version, Sept ’07) concluded:– JDEM should start in 2009 because LISA must wait for LPF– After JDEM, NASA should make LISA its top priority for

Beyond-Einstein goals. NASA should attempt to match ESA’s timetable for LISA launch in 2018.

– “Thus, the committee gave LISA its highest scientific ranking.”

gravitational waves: tool for astronomy

Documents

em waves

spectrumgravitational

taylorhulsegravitational

orbitgravitational waves

gravitational effect

launch of lisa

lisa pathfinder

lisa waveband