D - chgk.info

chgk.info

D - chgk.info

Prospects inGravitational-Wave Astronomy(NSF AAPF,(Image: MPI for Gravitational Physics / W.Benger-ZIB)Ilya Mandel@Northwestern University)May 14, 2010RIT Seminar


Gravitational-wave observatoriesLIGO:LaserInterferometerGravitational-waveObservatoryLISA:LaserInterferometerSpaceAntennaRIT: May 14, 20102


Gravitational WavesRipples in spacetime:Inspiral sound borrowedfrom Scott HughesCaused by time-varying mass quadrupole moment; GW frequencyis twice the orbital frequency for a circular, non-spinning binaryIndirectly detected by Hulse & Taylor [binary pulsar]Huge amounts of energy released: 5% of mass-energy of asupermassive black hole binary is comparable to theelectromagnetic radiation emitted from an entire galaxy over theage of the universe!RIT: May 14, 20103


Why do we want to see GWs?RIT: May 14, 20104


Why do we want to see GWs?RIT: May 14, 20105


Why do we want to see GWs?RIT: May 14, 20106


Types of GW sourcesContinuous sources [sources with a slowlyevolving frequency]: e.g., non-axisymmetricneutron stars, slowly evolving binariesCoalescence sources [known waveforms,matched filtering]: compact object binariesBurst events [unmodeled waveforms]: e.g.,asymmetric SN collapse, cosmic string cuspsStochastic GW background [early universe]??? [expect the unexpected]RIT: May 14, 20107


LIGO (Laser InterferometerGravitational-Wave Observatory)h ∼ ¨Qd ∼ mr2 ω 2∼ m M- 4 km long armsd d r- Typical strains h = ΔL /L ~ 10 -21 (NS-NS in Virgo cluster)- Needs to measure ΔL = hL ~ 10 -18 m- 2 LIGO detectors in US + Virgo, GEO in Europe- Virgo has 3 km baseline; data-sharing agreement with LIGORIT: May 14, 2010


Detection ChallengesRIT: May 14, 20109


Detection ChallengesRIT: May 14, 201010


LIGO Noise SpectrumRIT: May 14, 201011


Advanced LIGO- ~ x10 in range -> ~ x1000 in event rate- 10 Hz low frequency cutoffRIT: May 14, 201012


Rates predictions Ground-based interferometric detectors (LIGO, Virgo,GEO 600, AIGO, LCGT) are sensitive @ tens/hundredsHz: ideal for detecting NS-NS, NS-BH, BH-BH binaries Coalescence rate predictions from:» extrapolation from observed binary pulsars» simulations of isolated binary evolution» dynamical-formation models» intermediate-mass-black holes ? Instrument sensitivity and conversion to detection rates All astrophysical rates estimates depend on limitedobservations and/or models with many ill-understoodparameters, and are still significantly uncertain at presentRIT: May 14, 201013


Extrapolation from BNS observations Best NS-NS merger-rateestimates come fromobserved Galactic binarypulsars Small-number statistics(~10 total, ~5 merging in15 Gyr) Selection effects (pulsarluminosity distribution) [Kim et al., 2003 ApJ 584 985,2006 astro-ph/0608280;Kalogera et al., 2004, ApJ 601L179]RIT: May 14, 201014


Constraining models Add constraints from observations; binary pulsars: NS-NS,NS-WD, supernovae, etc. Average over models that satisfy constraintsRIT: May 14, 201016


Effect of adding constraints, 1Single constraint satisfaction - no accounting forsampling uncertainties or model fitting errorsRIT: May 14, 201017


Effect of adding constraints, 1Single constraint satisfaction - no accounting forsampling uncertainties or model fitting errorsRIT: May 14, 201017


Effect of adding constraints, 2Constraints fromobserved binary pulsarsBH-NS and NS-NSrate/MWEG predictions[OʼShaughnessy et al., 2008, ApJ 672 479]RIT: May 14, 201018


Merger and Detection RatesSource R low R re R plNS-NS (L −110 Myr−1 ) 0.6 60 600NS-BH (L −110 Myr−1 ) 0.03 2 60BH-BH (L −110 Myr−1 ) 0.006 0.2 20√S n(f ), 1/√Hz!" !#"!" !#!!" !##!" !#$+,-.-/0(1+23+,-.-/0(4-56789:/,;


Ṅ = R × N G(merger rate) =(merger rate per L10) *(Ng in L10's)ρ ≡√4∫ fISCO0ρ(D horizon ) ≡ 8LIGO sensitivity|˜h(f)| 2S n (f) df1/2.26 -- sky and orientation[Kopparapu et al., 2008 ApJ 675 1459 ]averaging; 0.02N G (L 10 )= 4 ( L10 per Mpc ) 3 33 π Dhorizon(2.26) −3 (0.02)MpcS4 S5 aLIGO|˜h(f)| =2/D ∗ (5µ/96) 1/2 (M/π 2 ) 1/3 f −7/6RIT: May 14, 201020


Dynamical Formation BH-BH mergers in dense black-hole subclusters of globularclusters» [OʼLeary, OʼShaughnessy, Rasio, 2007 PRD 76 061504]» Predicted rates 10 -4 to 1 per Mpc 3 per Myr» Plausible optimistic values could yield 0.5 events/year for Initial LIGO BH-BH scattering in galactic nuclei with a density cuspcaused by a massive black hole (MBH)» [OʼLeary, Kocsis, Loeb, 2009 arXiv:0807.2638]» Based on a number of optimistic assumptions» Predicted detection rates of 1 to 1000 per year for Advanced LIGO BH-BH mergers in nuclei of small galaxies without an MBH» [Miller and Lauburg, 2009 ApJ 692 917]» Predicted rates of a few X 0.1 per Myr per galaxy» Tens of detections per year with Advanced LIGORIT: May 14, 201021


Inspirals into IMBHs Intermediate-mass-ratio inspirals of compact objects(1.4 solar-mass NSs or 10 solar-mass BHs) intointermediate-mass black holes in globular clusters Dominant mechanism:IMBH swaps into binaries,3-body interactions tightenIMBH-CO binary, mergervia GW radiation reaction[IM et al., 2008 ApJ 681 1431] Rate per globular cluster: few x 10 -9 yr -1 Predicted Advanced LIGO event rates between 1/fewyears and ~30/yearRIT: May 14, 201022


Inspirals of two IMBHs Two very massive stars could form in globular clusterswith sufficient binary fraction, then grow through runawaycollision to form two IMBHs in same GC Rates of order 1/year arepossible for AdvancedLIGO [Fregeau et al., 2006ApJ 646 L135] IMBH binaries could alsoform when two GCs merge[Amaro-Seoane and Freitag,2006, ApJ 653 L53]RIT: May 14, 201023


Astrophysics with GW searchesConstraints on astrophysicalparameters from existingelectromagnetic observations[OʼShaughnessy et al., 2008 ApJ 672 479]:RIT: May 14, 201024


Astrophysics with GW searchesConstraints on astrophysicalparameters from existingelectromagnetic observations[OʼShaughnessy et al., 2008 ApJ 672 479]:Observed GW event rates can becompared with models to determineimportant astrophysical parameters;RIT: May 14, 201025


Astrophysics with GW searchesConstraints on astrophysicalparameters from existingelectromagnetic observations[OʼShaughnessy et al., 2008 ApJ 672 479]:Observed GW event rates can becompared with models to determineimportant astrophysical parameters;Could match measured massdistributions, etc. to models (requiresaccurate parameter determination)RIT: May 14, 201026


Markov Chain Monte CarloAnimation by Marc van der SluysRIT: May 14, 2010van der Sluys, IM, Raymond, et al., 0905.132327


Accurate Parameter Estimationvan der Sluys, IM, Raymond, et al., 0905.1323RIT: May 14, 201028


Astrophysics with GW searchesConstraints on astrophysicalparameters from existingelectromagnetic observations[OʼShaughnessy et al., 2008 ApJ 672 479]:Observed GW event rates can becompared with models to determineimportant astrophysical parameters;Could match measured massdistributions, etc. to models (requiresaccurate parameter determination)As detector sensitivity improves,even upper limits can be useful inconstraining parameter space forbirth kicks, common-envelopeefficiency, winds, etc.RIT: May 14, 201029


Astrophysics with GW searchesObserved GW event rates can be compared with models to determineimportant astrophysical parameters;Could match measured mass distributions, etc. to models (requiresaccurate parameter determination)Constraints from upper limits - example[IM & O’Shaughnessy, 2009]RIT: May 14, 2010 30


Astrophysics with GW searchesObserved GW event rates can be compared with models to determineimportant astrophysical parameters;Could match measured mass distributions, etc. to models (requiresaccurate parameter determination)Constraints from upper limits - example[IM & O’Shaughnessy, 2009]RIT: May 14, 2010 30


LISA:Laser Interferometer Space AntennaRIT: May 14, 20103 spacecraft around the Sun,5 million km apart31


LISA Binary Sources LIGO sensitive @ a few hundred Hz» NS-NS, NS-BH, BH-BH binaries LISA sensitive @ a few mHz» massive black-hole binaries– merger tree models to describe history of Galactic mergers– could be detected anywhere in Universe, SNR up to thousands– a few to tens of detections [e.g., Sesana et al., 2005]» galactic white dwarf (and compact object) binaries– 30 million in Galaxy, create noise foreground [Farmer & Phinney, 2003]– 20,000 resolvable» extreme-mass-ratio inspirals of WDs/NSs/BHs into SMBHs– complicated modeling of dynamics in Galactic centers: loss cone problem,resonant scattering, etc.– could see tens to hundreds to z~1 [e.g., Gair et al., 2004]RIT: May 14, 201032


Embarrassment of richesRIT: May 14, 2010[Arnaud et al., 2007, CQG 24 S551]33


LISA Data Analysish(t) =h(M 1 ,M 2 , ⃗ S 1 , ⃗ S 2 , θ, φ,D L , e, ...; t)17 parameters8)9%%%%%%%%%%:5;%%%%%%%%%%%%?=%!"#$%"#&%#'()#*+%,))-%#../01'2&")*3%MLDC 1! MLDC 2! MLCD 1B! MLDC 3!•! Verification !!•! Unknown, !isolated!•! Unknown, !interfering!•! Galaxy of !"3x10 6 !•! Isolated !! •! 4–6x, !"over Galaxy"and EMRIs!•! Isolated !!•! 4–6x, over Galaxy"and SMBHs!•! Verification !!•! Unknown, !isolated!•! Unknown, !confused!•! Galaxy !"of 6x10 7 "chirping!•! Isolated !! •! Over Galaxy !spinning,"precessing!•! Isolated !! •! 5 !"together,"weaker!•! Cosmic string !cusp bursts!•! Cosmological !background!Need innovative searchtechniques to separate manyoverlapping signals: Markov-Chain Monte Carlo,MultiNest, genetic algorithms4#,')%,+%56%7#''2&-)(2%Mock LISA Data Challenges[Gair, IM, Wen, 2008, CQG 25 184031]RIT: May 14, 201034


SMBH binariesfrom [Arun et al. (LISA Parameter EstimationTaskforce), 2008, CQG 26, 094027]RIT: May 14, 201035


Mock LISA Data Challenge Results!"#$$%&'%()*+(,(-#../0%(1$#23(45$%.(!"#$$%&'%()*+(,(-#../0%(1$#23(45$%.(-5&6%(!#7$5(89(:%/$(!57&/."((((((();(%7'%7(-5&6%(!#7$5(89(:%/$(!57&/."(;(%7'%7(RIT: May 14, 201036


Extreme Mass Ratio InspiralsSound fromScott HughesAnimationfrom Jon GairRIT: May 14, 201037


Exploring the spacetime...RIT: May 14, 201038


... taking lots of picturesRIT: May 14, 201039


Testing the “no-hair” theoremRIT: May 14, 201040


Testing the no-hair theoremRIT: May 14, 2010 41


Testing the no-hair theorem?Stationary, vacuum, asymptotically flat spacetimesin which the singularity is fully enclosed by ahorizon with no closed timelike curves outsidethe horizon are described by the Kerr metricRIT: May 14, 2010 42


Do black holes have hair?t traversed the neck more than once. Further adjustment of the orbital parameters causesneck to widen and eventually disappear. At that stage, most of the orbits appear to beular, but orbits that pass very close to the inner edge of the merged region (i.e., close toCTC zone) have not been fully investigated.An alternative explanation of these results [19] is that the geodesic equations are nuricallyunstable in the inner region, and therefore small numerical round-off errors in thegration routines are driving the orbits away from their true values. Once again, thistinction is not relevant observationally. An astrophysical system harboring an EMRI willbe isolated. The gravitational perturbations from distant stars etc. will serve the samein perturbing the orbits as numerical errors might on a computer. The end result —t the orbit is apparently ergodic — is the same.0.1540.130.052d!/d#0!0.05d!/d#10!0.1!1!0.15!2!0.24 5 6 7 8 9 10 11 12!/"FIG. 6: Poincare map for a geodesic in the outer region of Fig. 4.!30.8 1 1.2 1.4 1.6 1.8 2!/"[Gair, FIG. Li, 7: Poincare IM, 2009, map for a geodesic PRD in the77:024035]inner region of Fig. 4.RIT: May 14, 2010 43frequencies at a high level of precision (1 part in 10 7 for the first ∼ 10 harmonics). This


he future: 3rd-generation detectors The Einstein Telescope:» Underground, sensitive to 1 Hz» Exciting science example: mergersof light seeds of massive blackholes at high redshifts [Sesana,Gair, IM, Vecchio, 2009] ALIA/DECIGO/BBO» Space-based LISAs on steroids» Exciting science example: using300,000 merging binaries asstandard candles for precisioncosmology: Hubble constant to0.1%, w to 0.01 [Cutler & Holz, 2009] Pulsar timing» Sensitive to SMBHBs @ 10 -8 Hzfrom [Gair, IM, Sesana, Vecchio, 2009]RIT: May 14, 201044


Summary Current understanding of coalescence rates andproperties of compact binaries is imperfect Advanced LIGO is likely to see NS-NS, NS-BH, BH-BHcoalescences; tens or more coalescences may be seenaccording to some models, including dynamical formation Improved understanding of astrophysics can help GWsearch by informing detector configuration, template family GW detections and upper limits for compact-objectcoalescences will allow us to constrain the astrophysicalparameters Future GW detectors (LISA and beyond) will allow preciseprobes of a wide range of astrophysical environmentsRIT: May 14, 201045

More magazines by this user
Similar magazines