Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
3 Gravitational Wave Astronomy A major challenge and opportunity in relativistic gravitation is the direct detection of gravitational radiation from astrophysical sources. The expectation is that technology now under develop ment wiD within the next 10 years detect gravitational radiation or, if nature is insufficiently kind, that we will at least be able to set interesting astrophysical limits on the gravitational waves inci- dent on the Earth. These terrestrial observations will be made at gravitational wave frequencies above 10 Hz. The exploration of the potentially rich low-frequency spectrum of gravitational waves can only be carried out from space. The development of low-frequency gravitational wave detectors and ultimately a space gravitational wave observatory is a major new direction proposed in this study for the space program. The observation of gravitational radiation from astrophysical sources has several features. First, the direct detection of the waves will serve to test relativistic gravitation by measuring the propaga- tion speed and polarization states of the waves. The sources of the waves will most likely be regions in which the gravitational field strength is large. Thus, the detection of signals from these regions will serve to test gravitation in the strong-field, high-velocity limit. Second, gravitational radiation is very weakly coupled to matter and will not scatter even in the strongest sources. Observations 21
22 of phenomena deep In the interior of regions normally obscured in the electromagnetic astronomers will thus become accessible. The processes involved in stellar collapse and the pruneval cosmic kernel are two weD-known examples. Estimates of the gravitational wave flux incident on the Earth from astrophysical sources have been divided into three categories: sources of gravitational wave impulses or bursts, periodic sources that may produce continuous gravitational wave trains, and, fi- nally, sources of a gravitational wave background noise. The cat- egories are related to different techniques for detection. The es- tunates themselves are of varying quality. In some cases, such as the radiation from ordinary binary stellar systems, the estimate is reliable. Should no radiation be observed, it would indicate a failure of the theory. For other sources, such as supernova ex- plosions, the occurrence of the phenomena is wed established but our ignorance of the physical processes involves} leave the ampli- tude of the waves uncertain. Yet another class consists of posited sources whose number is unknown, but for which the amplitude of the waves is calculable; black holes are the best example of this class. Prediction in astrophysics is always a hazardous exercise, especially when one opens a new field or when a profound change in sensitivity makes new observations possible. The theoretical predictions of what might be discovered at the birth of x-ray as- tronomy certainly clid not anticipate the enormous diversity of the phenomena that subsequently were uncovered. Supernova explosions are known to occur In our galaxy at a rate of between 1 and 10 per century. The collapse of the core of a supernova to a neutron star or black hole could be accom- panied by the release of gravitational radiation if the collapse Is not spherically symmetric. The time scales of the collapse dur- ing which gravitational radiation would be emitted lie between 1 and 10 ms, but the fraction of the explosion energy going into gravitational radiation is uncertain. A supernova at the center of our galaxy that releases ~ percent of its energy into gravitational radiation would produce a wave amplitude at the Earth having a strain of 10~~. Present-day gravitational antennas would detect such a pube. In part, the goal of ground-based efforts to detect gravitational radiation has been set by the search for supernova events to achieve a strain sensitivity of 10-2i, which would observe these events to a distance of the Virgo cluster of galaxies at an event rate of 1 to 10 per year.
23 Pulsars are periodic sources whose number of occurrences is known but whose amplitude of gravitational radiation is uncertain. The amplitude depends on the mass eccentricity of the spinning star. Should the eccentricity be entirely due to distortion of the star by the magnetic fields trapped In the star during the col- lapse from an ordinary star, the wave amplitudes from pulsars in our own galaxy would give strains of 10-32 to 10-33, much too small to measure. However, the pulsar could have an intrinsic mass eccentricity, which, if it was as large as 10-5, would produce a measurable strain of 10-26. Earth-based detectors with wave frequencies above 10 Hz are now being planned with such sensitiv- ities. Many of the pulsars have frequencies around ~ Hz and thus would be candidates for space antennas. Ordinary binary stellar systems abound in the galaxy; approx- imately ~ percent of all stars are members of binary systems. The closer and fast ordinary binaries are sources of gravitational radi- ation at strain levels of 10-2~ with periods ranging between 1 to 10 h. These are clear candidates for detection by space antennas. In fact, there may be so many of them that they could constitute an unresolved stochastic background of gravitational waves. A subclass of binary stars includes the double neutron star systems such as PSR 1913+16. These are particularly interesting for both space- and ground-basec! gravitational antennas. PSR 1913+16 now radiates at submultiples of 8 h with strain amplitudes around 10-23, which would be detectable by a space antenna. After about one million years, as this system loses energy through gravitational radiation, it will produce a gravitational wave chirp that will be detectable by ground-based antennas. The system will then spend about 1 year near a period of 10 s and ultimately come to an abrupt end in 1 ms as the two stars collide. The gravitational wave strain multipliecl by the square root of the number of cycles lies in the vicinity of 1~)-~. Three binary neutron star systems from a total population of several hundred pulsars are known to exist in our galaxy. Extrapolating to the rest of the universe, we could expect to detect an event of this type every few hours in an antenna with a strain sensitivity of 10-22 to 10-23. Binary systems composed of double white dwarf stars are fairly certain to exist. These systems radiate at periods of 1000 to 100 s with strain amplitudes in the region 10-2° to 10-22. Not observable by ground-based antennas, they fit well into the best performance region of a projected space antenna.
24 There are a host of more speculative sources. The gravita- tional radiation from the collision of a black hole vnth another black hole or other compact object as well as the radiation emutted in the formation of a black hole is well studied. The radiation originates both from the acceleration of the masses and from the excitation of ringing in the normal modes of the metric solution around the black hole. The gravitational wave bursts from such events have large strain amplitudes with frequency components that vary as the inverse of the black hole masses. The formation of a l~solar-mass black hole at a distance of 100 megaparsecs (Mpc) could produce a strain pulse of 10-2t lasting ~ ms. Should black holes form binary systems, the orbital decay of a l~solar- mass black hole binary system anywhere in the universe could be observed with an antenna having a strain sensitivity of 10-22. The space antennas are well suited to measure the radiation from massive black holes. The formation of a 107-solar-mass black hole anywhere in the universe would produce a strain at periods of several hours of 10-~6 or larger. One of the most interesting speculative sources of gravitational radiation under current theoretical consideration Is the radiation suffusing the universe that may have originated In the universe's earliest epoch. Should present thinking be correct, it is possible that quantum gravitational wave fluctuations during the Planck epoch were amplified in the subsequent universal expansion. The radiation would appear as a gravitational wave background noise. The spectrum of the radiation is not well understood however, it is believed to contain less than 10-4 of the energy density required to spatially close the universe. The search for cosmic background of gravitational radiation is a prime motivation for both space- and ground-based gravitational wave antennas. The sensitivity of gravitational wave observations on the ground is advancing rapidly. Acoustic bar detectors at cryogenic temperature using low noise position transducers are now able to search for gravitational wave bursts In the kilohertz band with a strain sensitivity of 10~~. Several detectors are now operating in coincidence, and results of the searches at this level of sensitivity should be available in 1986. The acoustic detectors will continue to improve but must be able to circumvent the Enliven quantum limit, which will set in at strain sensitivities between 10-2° and
25 10-2i depending on the frequency of observation. Acoustic detec- tors will be constructed at lower frequency and as the transducer technology unproves could have bandwidths of around 10 percent. The other ground-based technique for detecting gravitational radiation utilizes laser interferometer systems that measure the separation of a configuration of free masses. This technique is also a precursor for the most promising space antennas. At present the largest of these systems are 30 to 40 m long. Using light powers of several hundred milliwatts, they can attain strain sen- sitivities of 10-~7 at 1 kHz. The promise for greatly enhanced sensitivity lies in constructing these systems with 100 times larger length and increased position sensitivity by increasing the light power modulated by the interferometer. The systems are ~nher- ently broadband, and as the techniques for reducing the stochastic forces on the masses unprove, they are expected to perform at a pube strain sensitivity of better than 10-22 between 10 Hz and 1 kHz. The sensitivity for periodic sources is expected to be less than 10-26 for integration times of a month. The present plan is to construct two ~km-Iong antennas in the United States, and there are plans in Great Britain, Germany, and trance to construct an- tennas of comparable length. These antennas will be operated as a network to determine the position of gravitational wave sources in the sky. These systems will be limited by ground noise and gravity gradient noise to operate above 10 Hz. The exploration of the gravity wave flux at longer periods is clearly the domain of space research, where longer baselines are possible and smaller low-frequency stochastic forces will be encountered.
