Sources of gravitational waves for groundbased detectors

1
Sources of gravitational waves for ground-based
detectors

• Ben Owen
• Penn State

2
Whats new since Amaldi-5?
(selected highlights)

• Stochastic
• Not much in this frequency band
• Burst
• Supernova simulations still dont explode, but
• Inspiral
• Newly observed binary pulsar increases event
  rates
• Periodic
• Several new theories predict big neutron star
  signals

3
Burst sources

• Short-lived, roughly modeled signals (cant make
  matched filters)
• Waveforms rough amplitude, band, duration
• Populations
• Gamma-ray bursts (NS/NS, BH/NS, hypernovae)
• Soft gamma repeaters (NS modes)
• Supernova core collapse
• Dynamical instabilities in proto-neutron stars
• Things that go bump in the night

4
Supernova core collapse

• Ott et al., ApJ 600, 834 (2004) Mueller et al.,
  ApJ 603, 221 (2004)
• Now realistic equations of state, neutrino
  transport
• Strain down by 4-10, broad frequency band
• Signal from post shock convection, modest spin
• LIGO-I detectability at 10kpc is 10 (angles)
• But they dont explode!

From Ott et al. 2004
5
Inspiral sources (binaries)

• Short signals modeled well enough for matched
  filtering (optimal signal-to-noise)
• Waveforms
• Analytical post-Newtonian for NS/NS
• Numerical relativity for BH/BH
• BH/NS in between
• Populations
• NS/NS observed, though only a few (one more now!)
• BH/NS and BH/BH from stellar evolution simulations

6
PSR J0737-3039

• Burgay et al., Nature 426, 531 (2003)
• 2.4hr orbit means 85Myr coalescence time
• Both neutron stars pulse 22.7ms and 2.8s
• Dominates statistics (7x old event rate)
• Factor 3.5 due to short lifetime (total 185Myr)
• Factor 2 from short orbit spin periods (Doppler
  smearing makes it hard to detect, must be more)
• Maybe more since its dim? Need statistics

7
Empirical (NS/NS) event rates

• Kalogera et al., ApJ 601, L179 (2004) erratum
• Bayesian probability distributions of rates
• Assume a population (luminosity distribution)
• Simulate how many pulsars radio surveys have
  found
• Time between detections
• Initial LIGO 20yr median, 95 confidence gt 8yr
• Adv. LIGO 2day median, 95 confidence gt 12hr

8
Population synthesis event rates

• OShaughnessy et al., astro-ph/0504479
• Now use distribution of populations
• For each population, evolve progenitors until
  NS/NS
• Keep only those populations consistent with
  observed type Ibc supernovae wide NS/NS
  binaries
• Results
• BH/NS 0.5 NS/NS (lack of BH/NS observations)
• BH/BH lt 5 NS/NS (BH masses 7Msun)
• And NS/NS down by 10 from previous slide!
• But distribution of populations is from outdated
  data

9
Waveforms

• Analytical
• Post-Newtonian is done for ground-based (3.5PN)
  Blanchet et al., PRL 93, 091101 (2004) errata
• Numerical
• First BH/BH orbit (or more) Bruegmann et al.,
  PRL 92, 211101 (2004)
• Waveform extraction remains a problem
• Some progress on realistic initial data
• New methods to suppress constraint violations

10
Periodic sources

• Long-lived, nearly monochromatic, data analysis
  dominated by Doppler shifts
• Population rotating neutron (or quark) stars
• Key number is ellipticity e (Ixx Iyy) / Izz
• Comparable to quadrupole/Izz, DR/R
• Standard max. e lt few x 10-7 for neutron star
  crust
• Detectability (initial interferometers)
• Known pulsars e lt 10-6 (distances few kpc)
• All-sky unknown pulsar search limited by
  computational cost range lt 1kpc for e lt10-5

11
Magnetic field bumps

• Cutler, PRD 66, 084025 (2002)
• Differential rotation makes B field toroidal
• Toroidal field pinches star
• Field axis moves toward equator, makes
  ellipticity
• Max. e lt few x 10-6
• But field needs to be buried (LMXBs and recycled
  pulsars?)

12
Magnetic bottling

• Melatos Payne, ApJ 623, 1044 (2005)
• Accreting ms pulsars
• Matter crosses field lines very slowly,
  accumulates at magnetic poles
• Depends on low-density conductivity (robust)
• Max. e 10-5 if mountain is wide (quadrupole)
• But e lt few x 10-7 for known stars from x-rays

13
Mountains (solid)

• Owen, astro-ph/0503399
• Mostly solid stars
• Max. e lt few x 10-4 (quarks)
• Max. e lt 10-5 (hybrid star)
• Uncertain physics of high density matter
• LIGO limits already there
• PRL 93, 181103 (2005) 9 new es within quark
  range
• But how to get max. e? (many pulsars are lt 10-8)

14
Other asymmetries

• R-modes (accreting stars)
• Duty cycle of emission was thought to be low
  (10-7)
• Wagoner, ApJ 578, L63 (2002) Andersson et al.,
  MNRAS 337, 1224 (2002) accreting stars are good
  sources if exotic particles increase bulk
  viscosity
• Free precession
• Was thought to damp in 1yr (vortex pinning)
• Van den Broeck, CQG 22, 1825 (2005) Observations
  say precession lasts gtgt10yr, visible to galactic
  core with advanced interferometers

15
Summary

• Theory EM observation both giving good news
• Recent upper limits are already (borderline)
  astrophysically interesting
• Initial detectors have a fighting chance (order
  10) of detecting something!
• (this is where Ladbrokes stopped taking bets)