Inner Workings: Spotting gravitational waves using pulsar ... · 7 Shannon RM, etal. (2015) Gravitational waves from binary supermassive black holes missing in pulsar observations.

INNER WORKINGS

Spotting gravitational waves using pulsar ticksAdam Mann, Science Writer

In 2003, a group of Japanese astronomers studyingthe center of galaxy 3C 66B thought they’d spotteda pair of record-breaking supermassive black holesorbiting one another: each of them appeared to havethe mass of more than 27 billion suns (1). But a dis-senting paper soon appeared, one that gleaned itsinsights from an unexpected source. A team watchinga pulsar—a rapidly rotating neutron star that shootsout a beam of radiation like a cosmic lighthouse—knew that such amassive black hole binary systemwouldbe emitting powerful gravitational waves, which wouldhave interferedwith the pulsar’s signal as they swept past(2). Hence, their observations suggested no such blackholes existed. Pulsar astrophysicist Andrea Lommen ofFranklin & Marshall College in Lancaster, Pennsylvania,remembers a conversation with collaborator Rick Jenet,an astronomer at the University of Texas at Brownsville.“Rick called me and said ‘They don’t have a source thatbig because we’d see it,’” recalls Lommen.

Until this point, many researchers had consideredthat it might one day be feasible to use pulsars todetect the presence or absence of gravitational waves.But few knew that the field had matured to any sort ofpractical applications. “I think that kind of woke peo-ple up and made them realize this pulsar-timing thingisn’t just a ruse,” says Lommen. “We’re actually goingto do something.”

Listening for GravityAbout four years later, Lommen, Jenet, and a few oftheir colleagues came together to form the NorthAmericanNanohertzObservatory for GravitationalWaves(NANOGrav), a collaboration that aims to turn the localuniverse into a gigantic gravity wave listening device.Using the Green Bank Telescope and Arecibo Observa-tory, the researchers monitor 54 pulsars in the hopes ofspotting minute variations in their beams that couldindicate ripples in the fabric of space-time. Along withsimilar projects around the world, NANOGrav could

Using the Green Bank Telescope and Arecibo Observatory (pictured), the NANOGrav project aims to monitor pulsars inthe hopes of spotting minute variations in their beams, suggesting ripples in the fabric of space-time. Image courtesy ofShutterstock/Dennis van de Water.

8878–8880 | PNAS | August 9, 2016 | vol. 113 | no. 32 www.pnas.org/cgi/doi/10.1073/pnas.1611117113

INNER

WO

RKIN

GS

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 10

, 202

1

one day reveal important information about the dynamicsof black holes, the formation of galaxies, and potentiallyeven more exotic phenomena.

Like electromagnetic waves, gravitational wavescome in a spectrum of different frequencies. So theseefforts are complementary to those of the Laser In-terferometer Gravitational-Wave Observatory (LIGO),which announced its first gravitational wave detection inFebruary of this year and another in June, but which issensitive to waves 11 orders-of-magnitude smaller thanthe ones NANOGrav intends to see (3, 4). The obser-vations require extreme precision, and it’s been difficultto find enough useful pulsars and to develop thedetection methods. The NANOGrav team thinks theycould spot gravitational waves in the next 5–10 years,although exactly when they’ll finally capture their elusivesignal remains an open question.

Wrinkles in Space-TimePulsars were first discovered in 1967, when astronomersJocelyn Bell Burnell and Antony Hewish detected reg-ular radio pulses coming from a spot on the night sky(5). It was eventually determined these were being emittedby the remnant core of a gigantic star that had gone su-pernova, leaving behind a highly magnetized neutron starwhose beam was repeatedly sweeping past the Earth.Within a decade, other researchers had cataloged morepulsars and begun realizing they might be used to test apart of Einstein’s Theory of Relativity, which states thatmassive accelerating objects radiate gravitational waves (6).

“If you think of the pulsar as a clock, it has an internalrotation period, which is very stable,” says astrophysicistMaura McLaughlin of West Virginia University, anothermember of the NANOGrav team. “If nothing else washappening, we’d see its pulses bump, bump, bump—perfectly regular.”

A gravitational wave is a wrinkle in the fabric ofspace-time. Should one come between the Earth anda pulsar, the distance between them would shrink orstretch and the pulsar’s pulse would arrive slightlysooner or later than expected. A single pulsar signalarriving off beat probably wouldn’t mean much. But ifscientists saw the exact same shift in pulsars all overthe sky, it could indicate a passing gravitationalwave. The fluctuations are extremely tiny; over thecourse of five years the method would notice devi-ations of just a few nanoseconds from what wasexpected. So NANOGrav and other pulsar-timingarrays only study highly stable millisecond pulsars,which can rotate nearly a thousand times per secondand whose pulse arrivals can be predicted onnanosecond timescales.

“There are signals we would recognize as beinglike a pure tone, sort of like a tuning fork,” says Jenet.“And then there are signals that sound more like ifyou’re listening to a radio with nothing on it; you’dget that snow or white noise that sounds like rain.”

The pure tone waves would come from the after-math of a galactic collision. Nearly every galaxy isthought to have a supermassive black hole in its cen-ter. When two galaxies crash and merge their centralblack holes will orbit one another, produce gravita-tional waves, and eventually combine into a singlebehemoth black hole. This sequence can take several

billion years and there remain many uncertainties inmodels explaining the underlying physics. Gravita-tional waves from such a system would tell astrono-mers about the black holes’ size and speed, andoptical telescopes could perform follow-up observa-tions to help them learn the details of the process.

Such a finding would require there to be a relativelyclose supermassive black hole binary emitting powerfulgravitational waves. So the NANOGrav team expects tosee a more likely signal first: echoes from the era ofgalaxy formation. Cosmological simulations suggest thatgalaxies started out small and then collided with oneanother to produce the larger spiral galaxies seen in thepresent day. Each collision would have also involved a

supermassive black hole merger and the sum total oftheir gravitational wave emissions “should produce anoverall crinkling of space-time,” says Lommen.

This background would be a complex waveformpattern but it would be embedded with important in-formation. By studying it carefully, researchers couldanswer many open cosmological questions, such aswhether bigger galaxies consumed little ones over timeor if galaxies of roughly the same size merged together.

Then there are the unexpected results that comefrom opening a new way of looking at the universe. “Wemay see signals from early universe inflation that wouldplace constraints on the Big Bang,” says McLaughlin. Or,he adds, they may see the absence or presence of cos-mic strings, very dense and thin objects that some the-orists believe evolved during the earliest fractions ofcosmic history.

Are We There Yet?Many researchers are hopeful NANOGrav will observe asignal soon. “It’s really just a matter of time and effort,”said physicist AlanWeinstein of the California Institute of

As an enormous passing gravitational wave sweeps past the Earth, it distortsthe fabric of space-time. The pulsating signals from nearby pulsars are alsoaltered, arriving slightly earlier or later than expected. By carefully trackingthese deviations, researchers can reveal the gravitational wave’s presence.Image courtesy of B. Saxton (NRAO/AUI/NSF).

We may see signals from early universe inflation thatwould place constraints on the Big Bang.

—Maura McLaughlin

Mann PNAS | August 9, 2016 | vol. 113 | no. 32 | 8879

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 10

, 202

1

Technology, a LIGO teammember who isn’t working onpulsar-timing arrays. The NANOGrav collaboration hasbeen adding about five new pulsars to their catalog peryear, improving their sensitivity steadily over time.

But even after carefully accounting for all potentialinterference (from, for example, the motions of theEarth around the sun) and clocking a pulsar’s rotationperiod out to 15 decimal places, the Parkes PulsarTiming Array (PPTA) project, an Australian groupsimilar to NANOGrav, found no evidence of gravita-tional waves (7). The findings, released last year,suggested that models of the galactic gravitationalwave background needed to be revisited.

“It could be that when these supermassive blackholes are merging they are producing less gravita-tional waves than we think,” says astronomer RyanShannon of the Commonwealth Scientific and Indus-trial Research Organisation and ICRAR-Curtin Univer-sity, and a member of PPTA. “Or it could be there arefewer than people had predicted.”

The NANOGrav collaboration disagreed, point-ing out that the PPTA findings are based on moni-toring only four pulsars. “They did a careful job,but the astrophysics is complicated and we stillthink detection could be within reach in the nextfew years,” says McLaughlin. In March of this year,members of the team produced a paper concludingthat they have an 80% chance of finding gravita-tional waves within a decade (8).

In either case, the field has come a long way sinceits earliest days. In addition to NANOGrav and PPTA,there is also the European Pulsar Timing Array, andall three collaborate and share data as part of an in-ternational consortium. Jenet says that 20 years ago,any limits on gravitational waves from pulsar timingwere not much better than hand waving. “But nowwe’ve gotten to a point where the theorists have togo back and scratch their heads and say ‘Hmm. . . wedidn’t see it here. So what’s wrong with our astro-physical models?’”

1 Sudou H, Iguchi S, Murata Y, Taniguchi Y (2003) Orbital motion in the radio galaxy 3C 66B: Evidence for a supermassive black holebinary. Science 300(5623):1263–1265.

2 Jenet F, et al. (2004) Constraining the properties of the proposed supermassive black hole system in 3C66B: Limits from pulsar timing.Astrophys J 606(2):799–803.

3 Abbott BP, et al.; LIGO Scientific Collaboration and Virgo Collaboration (2016) Observation of gravitational waves from a binary blackhole merger. Phys Rev Lett 116(6):061102.

4 Abbott BP, et al.; LIGO Scientific Collaboration and Virgo Collaboration (2016) Properties of the black hole binary merger GW150914.Phys Rev Lett 116(24):241102.

5 Hewish A, et al. (1968) Observation of a rapidly pulsating radio source. Nature 217(5130):709–713.6 Sazhin MV (1978) Opportunities for detecting ultralong gravitational waves. Sov Astron 22:36–38.7 Shannon RM, et al. (2015) Gravitational waves from binary supermassive black holes missing in pulsar observations. Science 349(6255):1522–1525.

8 Taylor S, et al. (2016) Are we there yet? Time to detection of nanohertz gravitational waves based on pulsar-timing array limits.Astrophys J Lett 819(1):L6.

8880 | www.pnas.org/cgi/doi/10.1073/pnas.1611117113 Mann

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 10

, 202

1