Pre-Phase A Report - Lisa - Nasa

24 Chapter 1 Scientific Objectives
have more secure models of these systems and will in addition be able to estimate the
distance to the binary from the gravitational-wave amplitude and the estimated masses.
The orientation angle of the plane of the orbit will be interesting if other orientationdependent
phenomena are observed, such as jets or optical/radio/X-ray polarisation. The
large majority of galactic binaries will not be known in advance, but can be located on
the sky from the frequency modulation that the motion of LISA produces in their signals,
and to some extent also from the amplitude modulation. This is discussed in Section 4.4 .
Neutron star binaries. The best-known two-neutron-star (NS-NS) binaryisthefamous
Hulse-Taylor binary pulsar PSR B1913+16, discovered in 1975 [17]. Its orbital period
of 7.68 hrs places it below the LISA band, but it is important to LISA as the best-studied
member of a class of binaries that should be important sources. The Hulse-Taylor binary
is decaying due to the loss of orbital energy to gravitational waves at exactly the rate
predicted by general relativity [6]. PSR B1913+16 will coalesce to a single star in 3×108 years. Two other very similar systems are known. By considering the selection effects
in the detection of such systems, a recent detailed study [18] arrived at a conservative
estimate of N ∼ 103 such systems in the Galaxy, formed at a rate of about one per 105 yr.
Theoretical calculations of binary evolution give a wide variety of estimates of the number
of such systems. Most of them [19, 20, 21] give rather higher rates than the observational
estimates. It is possible, therefore, that observations give a lower bound on the number
of such systems, but that some fraction of the population does not turn up in pulsar
surveys. It may be that not all neutron stars turn on as pulsars, or even that binaries like
PSR B1913+16 may be merely the long-period tail of a distribution of binaries that are
formed with periods as short as an hour and which decay so rapidly through the emission
of gravitational radiation that one would not expect to see any such systems in pulsar
surveys. In this case the formation rate could be as high as one per 3 000 yr, leading to
a total population of N ∼ 3×106 systems. Moreover, recent observations of the binary
pulsar PSR J1012+5307, whose companion is a white dwarf that is much younger than
the apparent age of the pulsar as estimated from is spin-down rate, have suggested that
millisecond pulsar spindown may overestimate the pulsar’s true age [22]. Since binary pulsars
tend to be millisecond pulsars, this could also raise the binary neutron-star birthrate.
For a recent overview of this subject, see [23].
Another indication of this population comes from gamma-ray bursts [24]. From optical
identifications of some recent bursts, it is now known that these events occur at immense
distances [25]. Although the events are not understood in detail, it seems that they could
involve coalescences of neutron stars with other neutron stars or with black holes. Such
events occur at the end of the gravitational-wave evolution of systems in the population
of binaries we are considering here. Estimates of the size of the population from
observations of gamma-ray bursts are consistent with the observational limits mentioned
above. For example, the estimates above suggest that there could be of order 104 neutronstar/neutron-star
coalescences per year out to a redshift of z = 1. About 1000 observable
bursts are thought to occur each year, but it seems probable that bursts are beamed, so
that the two rates would be consistent for a 10 % beaming factor. If bursts are spread to
greater distances (one has been seen beyond z = 3), the rates are not consistent unless the
beams are very narrow, or unless the more distant bursts come from neutron-star/blackhole
mergers (see the next section), which could indeed emit stronger bursts, according
3-3-1999 9:33 Corrected version 2.08