\magnification=\magstep1
\parskip=\baselineskip
\centerline{\bf MATTERS OF GRAVITY}
\bigskip
\bigskip
\line{Volume 1, Number 1 \hfill Fall 1991}
\bigskip
\bigskip
\bigskip
\centerline{\bf Table of Contents}
\bigskip
\hbox to 6truein{Editorial {\dotfill} 1}
\hbox to 6truein{Correspondents {\dotfill} 3}
\hbox to 6truein{Experimental Notes {\dotfill} 4}
\hbox to 6truein{Completeness of Outgoing Stellar Modes {\dotfill} 5}
\hbox to 6truein{LIGO Project Report: October 1991 {\dotfill} 7}
\hbox to 6truein{Computational Relativity Grand Challenge {\dotfill} 8}
\hbox to 6truein{NSF Support for Gravitational Physics {\dotfill} 10}
\bigskip
\bigskip
\bigskip
\centerline{\bf Editorial}
\medskip
\centerline{Peter R. Saulson, Syracuse University}
\bigskip
Welcome to the debut issue of {\it Matters of Gravity}, a
newsletter for the gravitational physics community of the United States.
Gravitational physicists are a rare and diverse lot. Spread
around the country (and the world), we work mostly in small groups, with
a great range of technical approaches to our work. We would probably
benefit, both intellectually and politically, from a greater sense
that we are, in fact, a community. (It was only this summer that
the Physical Review recognized gravitational physics as a distinct area of
study by modifying the subtitle of the D15 issue. Thanks to Jim Hartle
for promoting this change.)
This newsletter is an attempt to help build that missing sense
of community. It grew out of informal discussions between Abhay Ashtekar and
a number of other attendees at the MG6 meeting in Kyoto last July. Our plan is
to produce {\it Matters of Gravity} four times a year. I've agreed to serve as
editor for the first year. Ashtekar will keep a hand in as associate
editor, with special responsibility for theoretical gravitation.
As you can see in the rest of this issue, the central feature of the
newsletter will be brief (one paragraph to one page) reports, summarizing
new scientific results, research initiatives, or political events relevant
to our work. We hope they will give a valuable overview of our diverse
field.
To make sure our coverage is as comprehensive as possible, we've
asked a number of you to serve as correspondents, each with responsibility for
one part of the field. Some will write reports themselves, others
will be contacting some of you to do the writing. The correspondents are listed
below. If you see an important topic that is not being well-covered, or would
like to volunteer your services as a correspondent, please contact the editor.
If you would like to submit a piece for the newsletter, you are welcome
to do so. You may either contact the appropriate correspondent, or
write directly to the editor. We accept submissions by e-mail, surface mail,
or fax, at the addresses given below. Eventually we will produce a style
manual for \TeX \ submissions via e-mail, but for now any legible format will
do.
This issue has been distributed in two forms. A paper edition,
produced in \TeX, was mailed to a list we derived mainly from the
NSF's list of sponsored workers in gravitational physics. A larger number
of you are receiving {\it Matters of Gravity} via e-mail.
(Invaluable assistance in electronic circulation
is being provided by Jorge Pullin.)
We hope that you will help us to reach people not on our initial list
who would like to receive a copy. (Or, indeed, if you would prefer not to
remain on our list, we will be pleased to strike your name.) Unless we hear
otherwise, we'll continue to send you the newsletter in the form you first
received it. A third format, e-mail distribution of a file with the
\TeX \ commands stripped out, will be made available if there is interest.
Please send requests for inclusion in or deletion from our subscription list to
the editor at one of the addresses listed below.
In future issues, we are considering also running announcements
of upcoming meetings or perhaps listings of available positions. In addition,
we would like your opinions on whether American gravitational physicists
should form some sort of organization, perhaps a national section of the
International Committee on General Relativity and Gravitation, or perhaps
an independent group. As in the recent past, there may be times in the
future when a quick and unified response from our community would be important.
Whether you agree or not, we would be interested in your comments.
Please also send me your comments on this first issue of the
newsletter, as well as suggestions for articles.
\medskip
\leftline{Peter Saulson}
\smallskip
\leftline{Department of Physics}
\leftline{Syracuse University}
\leftline{Syracuse, New York 13244-1130}
\smallskip
\leftline{Phone (315)443-5994}
\leftline{Fax (315)443-9103}
\leftline{Bitnet: saulson@suhep}
\leftline{Internet: saulson@suhep.phy.syr.edu}
\vfill
\eject
\centerline{\bf Correspondents}
\medskip
\item{1.} John Friedman and Kip Thorne: Relativistic Astrophysics,
\item{2.} Jim Hartle: Quantum Cosmology and Related Topoics
\item{3.} Gary Horowitz: Interface with Mathematical High Energy Physics,
including String Theory
\item{4.} Richard Isaacson: News from NSF
\item{5.} Richard Matzner: Numerical Relativity
\item{6.} Ted Newman: Mathematical Relativity
\item{7.} Bernie Schutz: News From Europe
\item{8.} Lee Smolin: Quantum Gravity
\item{9.} Cliff Will: Confrontation of Theory with Experiment
\item{10.} Peter Bender: Space Experiments
\item{11.} Riley Newman: Laboratory Experiments
\item{12.} Peter Michelson: Resonant Mass Gravitational Wave Detectors
\item{13.} Robbie Vogt: LIGO Project
\item{14.} Francis Everitt: Gravity Probe-B
\vfill
\eject
\centerline{\bf Experimental Notes}
\medskip
\centerline{Clifford M. Will, Washington University}
\bigskip
\bigskip
\leftline{\it New constraints for Moffat's Nonsymmetric Gravitation Theory}
Two recent papers have created problems for John Moffat's NGT. Gabriel {\it et
al.} ({\it Phys. Rev. D}, submitted) have shown that the violation of the
Einstein Equivalence Principle that is inherent in theories with a
non-symmetric metric causes light with different polarizations to propagate
differently in a gravitational field, and can lead to substantial
depolarization of light from, say, a magnetically active region on the solar
surface. Polarization measurements then place a strong constraint on the
variable ``l''-parameter of NGT. The other paper (Khaliullin {\it et al.,
Astrophys. J.} {\bf 375}, 314 1991) would seem to kick one of the legs out from
under NGT. It explains the anomalously small inferred periastron advance of
the binary system DI Her (which had been cited as evidence against general
relativity and in favor of NGT) as being caused by a third, distant body. The
effect is a combination of induced periastron shift, AND a periodic change in
the orbital eccentricity of DI Her, which changes how one translates the
observed eclipse times into a periastron advance.
\medskip
\leftline{\it A New Measurement of the Solar Gravitational Redshift}
LoPresto {\it et al. (Astrophys. J.} {\bf 376}, 757, 1991) have reported a
measurement of the gravitational redshift of chromospheric oxygen
lines in the Sun, with a value $0.99\pm 0.02$ of the prediction of the
equivalence principle.
\medskip
\leftline{\it VLBI Leads to Improved Light Deflection Measurements}
Using Very Long Baseline Interferometry, Robertson {\it et al. (Nature} {\bf
349}, 768, 1991) have reported measuring the deflection of radio waves by the
Sun, in agreement with general relativity at the 0.1 percent level. The
analysis used almost 350,000 observations of 74 radio sources, spread over the
entire celestial sphere. They can now see curved space everywhere! Truehaft
and Lowe ({\it Astronomical J.}, in press) have reported a VLBI measurement of
the deflection of light caused by Jupiter (300 microarcseconds) at about 50
percent accuracy.
\medskip
\leftline{\it Orbital Decay of the Binary Pulsar Agrees with General Relativity}
Observations of the binary pulsar during the summer of 1990 (J. Weisberg,
private communication), together with a small correction to the observed rate
of change of orbital period caused by galactic acceleration (Damour and Taylor,
{\it Astrophys. J.} {\bf 366}, 501, 1991) now imply that the observed rate is
$1.003 \pm 0.008$ of the general relativistic gravity-wave damping prediction.
The new binary pulsar PSR 1534+12 (Wolszczan, {\it Nature} {\bf 350}, 688,
1991) may ultimately lead to even better accuracy.
\vfill
\eject
\centerline{\bf Completeness of outgoing stellar modes}
\medskip
\centerline{John L. Friedman, University of Wisconsin, Milwaukee}
\bigskip
How can the outgoing modes of a star be complete in general
relativity? For Newtonian stars, a superposition of normal modes
describes an arbitrary perturbation of the fluid, but in relativity,
one has both fluid and field degrees of freedom. Outgoing modes
clearly cannot reproduce the incoming waves of an arbitrary radiation
field. Worse, all outgoing modes of a stable star blow up
exponentially at spatial infinity, and at first sight it appears that
they cannot model any regular perturbation of a star.
Along a future light cone, however, one can specify smooth,
finite-energy data for a normal mode, and it is possible to state two
natural completeness conjectures. Remarkably, Viqar Husain and Richard
Price [1] have just found a toy model of a relativistic star for which
both conjectures can be made precise, for which both turn out to be
true, and for which the proof is intuitively obvious. The conjectures
are [see also 2]:\break
(i) the normal modes are complete for initial data for the perturbed
fluid at t=0, and \break
(ii) the normal modes are complete in the space of purely outgoing
solutions to the Einstein-perfect-fluid equations restricted to the
causal future of the fluid at t=0.
The Husain-Price model of a relativistic star is a semi-infinite
spring (they call it a torsion bar), extending in 1-dimension from $r=0$
to $r=\infty$ (It is very close to a model considered earlier by
Kokkotas and Schutz (1986) [3]). The spring has two parts, with
different spring constants: A finite segment, the ``star", extends from
$r=0$ to $r=L$ and has wave velocity v, the speed of ``sound". The
remaining infinite part, extends from $r=L$ to $r=\infty$ and on it
waves have velocity c, the speed of ``light".
Because the wave operator with outgoing boundary conditions is not
self-adjoint, one would not ordinarily know how to prove completeness of
the normal modes. What saves the day is that for this particular
system, every outgoing solution $\phi$ is damped with the {\it same}
damping time $\tau$ and $\phi e^{it/\tau}$ is periodic. It is easy to
see why this has to be true.
The general solution in the ``star" is the sum of a right moving and a
left moving wave, $\phi = \phi_R(x-vt) + \phi_L(x-vt).$ Now the
left-moving wave, say, will reflect off the fixed origin, and after
time $L/v$ it return to its original position as a right moving wave
with its sign changed. It then propagates to the right, part of the
wave transmitted at $r = L$ and part reflected, with a reflection
coefficient $e^{-T/\tau},$ that is independent of frequency. Finally
after time $T = 2L/v$, it has returned to its initial position and
velocity, with its amplitude changed by the factor $(-1)(e^{-t/\tau})$,
as claimed. The right-moving wave, of course, performs the same
reflections in reverse order. Thus $\phi = \phi_L + \phi_R$ has the
property that $\phi e^{i(1+\pi)t/\tau}$ is periodic with period $T$.
Any function periodic in time with period $2L/v$ can be written as a
superposition of modes with frequency $n\pi v/L$, and any outgoing
solution $\phi$ can therefore be written as a sum of modes with complex
frequencies. Thus conjecture (i) is true: The outgoing normal modes are
complete in the space of solutions that are purely outgoing for $r>L$.
Conjecture (ii) is the statement that the outgoing modes
are complete for initial data on the segment $[0,L]$. This
is also true, because arbitrary initial data on [0,L] gives
rise to a unique outgoing solution on the spacetime.
Finally, an interesting result of Chandrasekhar and Ferrari [4] shows
that, although the outgoing modes of a star may be complete for the
fluid's initial data, they can carry information not
contained in the fluid variables. In considering the outgoing modes of
a dense spherical star, Chandrasekhar and Ferrari find an odd-parity
outgoing mode. Because odd-parity perturbations do not couple to a
spherical fluid, an odd-parity mode is unrelated to the fluid's degrees
of freedom. Its character is closer to that of the outgoing modes of a
black hole. Note that with a suitable definition of outgoing radiation,
black hole normal modes may be similarly complete for the set of
perturbations that are purely outgoing at infinity and ingoing at the
horizon.
\medskip
\leftline{REFERENCES}
\noindent [1] V. Husain and R.H. Price, ``A model for the completeness of
quasinormal modes of relativistic stellar oscillations.'' Submitted to
{\it Phys. Rev. Letters.}
\noindent [2] J. L. Friedman, ``Scenes from general-relativistic astrophysics,"
in {\it Recent Advances in General Relativity}, ed. A. Janis and J. Porter,
Birkauer, in press.
\noindent [3] K. D. Kokkotas and B. F. Schutz, {\it Gen. Rel. Grav.} {\bf 18},
913 (1986).
\noindent [4] S. Chandrasekhar and V. Ferrari, {\it Proc. Roy. Soc. A}, {\bf
434}, 449 (1991).
\vfill
\eject
\def\ltorder{\mathrel{\raise .3ex\hbox{$<$}\mkern -14MU\lower
0.6ex\hbox{$\sim$}}}
\centerline{\bf LIGO Project Report: October 1991}
\smallskip
\centerline{Rochus E. Vogt, Director, LIGO Project}
\bigskip
The objective of the Laser Interferometer Gravitational Wave Observatory (LIGO)
project is to establish a two-facility observatory to permit broadband
observations (10 Hz to 10 kHz) of gravitational waves at strain sensitivities
of $h \ltorder 10^{-23}$ for burst signals. This strain sensitivity may not be
obtained with the initial detectors, but the facilities are designed to permit
a sequence of detectors of higher performance to be developed and installed
with minimal additional cost. The ultimate sensitivity has been set to ensure
detection of coalescing neutron-star binaries (the most calculable source in
terms of strength and frequency of occurrence), but earlier detection of less
well understood sources (black-hole binary coalescence, black-hole formation,
supernovae) is quite probable.
The project is being pursued, under NSF sponsorship, by a team of scientists
and engineers from Caltech and MIT, with Caltech having fiduciary
responsibility for the project. LIGO will ultimately be operated as a national
facility open to the scientific community, offering opportunities for detector
development and data analysis. At present, formal collaborations have been
established with groups at Stanford University, the Joint Institute for
Laboratory Astrophysics, and Syracuse University. ~~LIGO will be part of an
international network forming a global observatory.
After undergoing a number of peer reviews, LIGO was approved by the National
Science Board in early 1990. Although proposed as a new start by President
Bush for FY'91, Congress only approved funds for continued design and R\&D.
~~LIGO again is in the President's budget as a new start in FY'92, at a level
of \$23.5M out of a 5-year construction total of $\sim$ \$211M. The House
Appropriations action reduced LIGO funding for FY'92 to \$0.5M, while the
subsequent Senate action allocated full funding of \$23.5M. On September 26,
the Senate/House conference committee apparently approved full funding
(\$23.5M) of LIGO for FY'92, and thus a construction start. We are awaiting
further details and the President's signing of the final bill after the full
Senate and House action.
Under a process approved by the National Science Board, the LIGO project opened
a national competition for the two LIGO sites. Eighteen site candidates from
17 states have been offered, and now are undergoing technical evaluation. NSF
action in the site selection process may occur by end of 1991.
R\&D on LIGO detectors continues under a number of initiatives at both Caltech
and MIT. Design of the first detectors to be installed in LIGO is underway.
Large-scale demonstration projects, verifying LIGO design concepts and
technology also are underway. These may be discussed in future contributions
to this newsletter.
For anyone who would like to learn more about LIGO, a writeup of an MG6 paper:
``The U.S. LIGO Project" (by R. Vogt) is available upon request to the LIGO
Project office, Caltech, M/S 102-33, Pasadena, CA 91125, or by e-mail to
information@ligo.caltech.edu.
\vfill
\eject
\centerline{\bf Computational Relativity Grand Challenge}
\smallskip
\centerline{Richard Matzner, University of Texas at Austin}
\bigskip
As a supplement to President Bush's Fiscal Year 1992 budget, the
Office of Science and Technology produced a report: {\it Grand Challenges:
High Performance Computing and Communication} (also known as the FCCSET
report, since it was prepared by OST's Federal Coordinating Council for
Science, Engineering and Technology.) This report proposes goals and strategies
to bring the art of computing to the Tera operation/sec level, and
networking to the 10-Gigabit/sec level by the year 2000. The strategy is to
support hardware, network, and {\it software and algorithm} development, by
involving different fields of numerical science in efforts that push the limits
of current resources. Emphasis will be placed on projects that truly require the
targetted computing power in order to succeed. Among the federal agencies
charged with supporting this effort is the National Science Foundation, which
funds a number of areas of computational physics.
Numerical Relativity is one of the most computationally demanding
numerical sciences. A collaboration has been formed to propose a Grand
Challenge effort in gravitational physics. The idea is to perform very
high accuracy computations of the two-body problem in general relativity,
and to predict the gravitational waveforms that result. At present, the
collaboration members are the University of Illinois (National Center for
Supercomputing Applications), Cornell University, the University of Pittsburgh,
the University of Texas at Austin (Center for Relativity), Northwestern
University, and the University of North Carolina at Chapel Hill.
We want to be able to solve generic problems without special
symmetries, that is problems which are fully three-dimensional. The simplest
such two-body problem is the two black hole case, since it removes all
complications of astrophysics. And black holes, with the strongest possible
gravitational fields, should be intrinsically strong and distinctive radiation
sources.
About a dozen Ph.D.s, and a larger number of students, are
already tackling different aspects of the problem. For instance, physically
does it matter if we evolve matter to collapse to black holes which then
collide, rather than taking ``eternal" black holes which have no astrophysics
and thus in some sense are simpler? Algorithmically is it better to take the
ADM variables, or use the Ashtekar-variable formalism? Should one evolve the
system by repeatedly computing space at ``one instant of time"? Or would a
null scheme be better, where one ``time" is in fact a whole outgoing null
cone, the entire history of radiation reaching infinity?
General Relativity allows wide latitude in coordinate choices.
Which are better? Curvilinear coordinates can better snuggle up to the curved
hole. But they tend to have difficult ``coordinate singularity" behavior,
and special points which require special coding attention. Rectangular
coordinates have no special points, and are easy to code in, but don't
conform very well to the structure of the problem. Many other such questions
must be considered. In some cases the solutions may be straightforward, but
careful attention must be given to each of these questions.
Estimates of the accuracy needed to evolve colliding black holes
accurately to produce radiation suggest zoning sizes like (1000)$^3$ spatial
zones, evolved at least 1000 steps into the future. Based on existing codes
which take about $2\times 10^{-5}$ seconds per timestep on Cray Y-MP computers,
this leads to years per simulation. (A Y-MP is capable of $2\times 10^8$
floating point computations per second.) Parallelism is an obvious direction
of improvement. The expectation is that well before the year 2000, parallel
computers now being designed will achieve $10^{12}$ floating point operations
per second, bringing the time scale for such a computation to the level of
hours.
In the process of developing the general relativity codes, we will
have tested and applied a wide range of numerical algorithms, and provided
an education in computational science to many graduate and undergraduate
students. We will also have produced a substantial library of community
computer codes. And we will have predicted the signals in the new generation
of gravitational wave detectors, which should allow us to probe to the edge of
the universe.
\vfill
\eject
\centerline{\bf Recent NSF Support for Gravitational Physics}
\smallskip
\centerline{Richard Isaacson, National Science Foundation}
\medskip
Present and recent NSF support for gravitational physics is
summarized in the two tables given below.
\bigskip
\centerline{\bf HUMAN RESOURCES}
\medskip
$$\vbox{\rm
\halign{\hfil #&\quad#\hfil &\quad\hfil # &\quad\hfil # &\quad\hfil # &
\quad\hfil # &\quad\hfil # \cr
& & FY87& FY88& FY89 & FY90& FY91\cr
\multispan7 \cr
\multispan7 \bf I. THEORY \hfil \cr
\multispan7 \cr
A.& \bf Analytic &&&&& \cr
1.& Faculty & 49 & 41 & 47 & 49 & 45 \cr
2.& Postdoc & 18 & 18 & 17 & 17 & 17 \cr
3.& Grad. Students & 17 & 18 & 18 & 15 & 15 \cr
4.& Undergraduates & --- & --- & 1 & 1 & 1 \cr
\multispan7 \cr
B.& \bf Computational &&&&& \cr
1.& Faculty & 7 & 6 & 7 & 10 & 10 \cr
2.& Postdoc & 3 & 3 & 3 & 6 & 8 \cr
3.& Grad. Students & 5 & 7 & 6 & 6 & 10 \cr
4.& Undergraduates & --- & --- & 1 & 2 & 1 \cr
\multispan7 \cr
C.&\bf Overall &&&&& \cr
1.& Faculty & 56 & 47 & 54 & 59 & 55 \cr
2.& Postdoc & 21 & 21 & 20 & 23 & 25 \cr
3.& Grad. Students & 22 & 25 & 24 & 21 & 25 \cr
4.& Undergraduates & --- & --- & 2 & 3 & 2 \cr
\multispan7 \cr
\multispan7 \bf II. EXPERIMENT \hfil \cr
\multispan7 \cr
1.& Faculty & 21 & 22 & 22 & 23 & 21\cr
2.& Postdoc & 9 & 9 & 7 & 11 & 13\cr
3.& Grad. Students & 19 & 21 & 25 & 25 & 24\cr
4.& Undergraduates & --- & --- & 15 & 16 & 14\cr
\multispan7 \cr
\multispan7 {\bf III. LIGO} (Caltech/MIT, Stanford, Colorado, Syracuse)\hfil \cr
\multispan7 \cr
1.& Faculty & 3 & 3 & 6 & 7 & 10\cr
2.& Postdoc & 11 & 11 & 10 & 8 & 11\cr
3.& Grad. Students & 8 & 9 & 7 & 11 & 14\cr
4.& Undergraduates & --- & --- & 6 & 10 & 10\cr} } $$
\vfill
\eject
\centerline{\bf GRAVITATIONAL PHYSICS}
\medskip
\centerline{\bf SUPPORT LEVELS}
\medskip
\centerline{{\bf AWARD SIZE} (ANNUAL RATE): \$ K/faculty}
$$\vbox{\rm
\halign{\hfil #&\quad#\hfil &\quad\hfil # &\quad\hfil # &\quad\hfil # &
\quad\hfil # &\quad\hfil # \cr
& & FY87& FY88& FY89 & FY90& FY91\cr
\bf I.& \bf THEORY &&&&& \cr
\multispan7 \cr
A.& \bf Analytic & 50.1 & 54.2 & 54.6 & 53.3 & 58.3 \cr
B.& \bf Computational & 69.5 & 83.7 & 71.2 & 66.8 & 80.4 \cr
C.&\bf Overall & 52.6 & 57.9 & 56.7 & 55.5 & 62.3 \cr
\multispan7 \cr
\bf II.& \bf EXPERIMENT & 112.0 & 114.8 & 121.7 & 116.2 & 124.5 \cr
\multispan7 \cr
\bf III.& \bf LIGO & 1065.6 &1199.8 & 669.1 & 598.5 & 516.8 \cr }} $$
\bigskip
\centerline{{\bf TOTAL FUNDING} (ANNUAL RATE) : \$ K }
$$\vbox{\rm
\halign{\hfil #&\quad#\hfil &\quad\hfil # &\quad\hfil # &\quad\hfil # &
\quad\hfil # &\quad\hfil # \cr
& & FY87& FY88& FY89 & FY90& FY91\cr
\bf I.& \bf THEORY &&&&& \cr
\multispan7 \cr
A.& \bf Analytic & 2407.0 & 2220.3 & 2565.5 & 2609.5 & 2623.7 \cr
B.& \bf Computational & 486.3 & 501.9 & 498.3 & 667.6 & 803.8 \cr
C.&\bf Overall & 2893.3 & 2722.2 & 3063.8 & 3277.1 & 3427.5 \cr
\multispan7 \cr
\bf II.& \bf EXPERIMENT & 2542.2 & 2525.6 & 2677.8 & 2673.3 & 2614.6 \cr
\multispan7 \cr
\bf III.& \bf LIGO & 3196.7 & 3599.3 & 4014.6 & 4189.8 & 5168.0 \cr
\multispan7 \cr
\bf IV.& \bf TOTAL & 8632.2 & 8847.1 & 9756.2 & 10140.2 & 11210.1 \cr} } $$
\bigskip
\centerline{\bf COMPOUND GROWTH RATES}
$$\vbox{\rm
\halign{\hfil # & \bf # \hfil & \hfil # \hfil \cr
&& ${\rm (FY91/FY87)^{1/4}}$\cr
\multispan3 \cr
I.& THEORY & 1.043 \cr
II.& EXPERIMENT & 1.007 \cr
III.& LIGO & 1.128 \cr
IV.& TOTAL & 1.068 \cr}}$$
\bye