US astronomy: Is the next big thing too big?

© 2006 Nature Publishing Group

Every ten years, US astronomers gettogether to list the things they wantmost in all the world — and outside it.This may sound like a grown up ver-

sion of writing to Santa Claus, but theseDecadal Surveys are taken very seriously. Pre-senting a united front on what matters most toone’s profession is a powerful bargaining toolwhen projects come up for political approval.

On astronomers’ most recent wish list, puttogether in 2000, pride of place was given towhat was then known as the Next GenerationSpace Telescope, an observatory that wouldtake up the mantle of the Hubble Space Tele-scope as Earth’s orbit-ing eye on the cosmos.Half a decade on, thattelescope, now namedafter former NASAadministrator JamesWebb, is well underway. But, as always, there’s a catch. At the begin-ning of the millennium, US astronomersthought that their most-wanted project wouldcost $1 billion. Its projected cost is now nearlyfive times that.

Price tags that mimic the Big Bang’s inflationare nothing new to astronomy. The problem forthe James Webb Space Telescope (JWST) is thatthe budgetary space in which it’s expanding isshrinking. Money once slated for science isbeing diverted to the space shuttle, the Interna-tional Space Station and plans for future

manned exploration (see Nature 439, 768–769;2006). So the costs of the Webb telescope areleading to cancellations — or ‘indefinite defer-rals’, as NASA prefers to call them. For thosewhose dreams are crushed in this process, theWebb telescope is looking less like the future oftheir field and more like its foreclosure. Onecritic of the process is Shri Kulkarni, anastronomer at the California Institute of Tech-nology in Pasadena. “My worry,” he says, “is thatwe are starting on a project whose cost we don’tunderstand, and which is now devouring space-based astronomy. I doubt there’s any projectthat is worth abandoning the rest of the field.”

Sterl Phinney is anastrophysicist, also atthe California Insti-tute of Technology,whose favoured pro-ject, the LISA gravita-tional wave telescope,

has just been deferred indefinitely by NASA.He says that even before the recent cuts, theWebb telescope was “basically sucking up allthe other money” astronomers hoped to use.The community is now split between thosewho view the situation with growing alarmand those who, according to Phinney, “reallylike the JWST and think it’s OK that it eatseverything else — although even some ofthose are worried about the balance and healthof astronomy.” Kulkarni thinks some in this second camp are still in “stunned shock”

over the most recent shift of funds away fromscience at NASA, and just haven’t reached aconsensus on what to say, let alone do, about it.

One thing on which everyone agrees is that,if it works as advertised, the Webb telescopewill be one fantastic machine. The telescope’s25-square-metre mirror is not just much big-ger than Hubble’s; it is bigger than any youwould have found at any observatory in theworld when Hubble launched.

The long viewHubble’s design is optimized for visible andultraviolet light, but the Webb telescope willsee in the infrared. Sitting above the atmos-phere, it will have an unfiltered view of aswathe of wavelengths from 0.6 �m (at the redend of the visible spectrum) all the way to thefirst fringes of the far infrared at 28 �m. Atlonger wavelengths, images of a given resolu-tion require a larger mirror; the Webb tele-scope’s honeycomb of burnished berylliumwill give it a resolution in the infrared that is assharp as Hubble’s is in the visible. The mirror’ssize also makes the telescope particularly sen-sitive: its instruments should see objects 10 to100 times fainter than Hubble can.

Going into the infrared means the telescopehas to have a big mirror and has to be sta-tioned far from Earth (the heat from whichwould otherwise be a problem). It also has tobe thoroughly shielded from the Sun, with astructure that somewhat resembles a multi-

NEWS FEATURE NATURE|Vol 440|9 March 2006

140

Is the nextbig thingtoo big?The cost of the James Webb Space Telescope could cripple US astronomy. Tony Reichhardt takes a closer look.

“We are starting on a projectwhose cost we don’t understand,which is devouring astronomy.”

— Shri Kulkarni

NO

RTH

ROP

GRU

MM

AN

9.3 space telescope MH 7/3/06 11:58 AM Page 140

Nature Publishing Group ©2006


NATURE|Vol 440|9 March 2006 NEWS FEATURE

141

storey trampoline. These requirements haveall driven up the telescope’s cost. But seeing inthe infrared is not an optional extra; it’s anecessity. If you want to look at the early Uni-verse, the infrared is where the action is.

Theory holds that after the glow of the BigBang faded, the Universe entered a long, light-less ‘dark age’. Eventually, knots in the cold darkmaterial condensed, collapsed and began toshine — the first stars. These earliest stars arereceding from us at a great rate, which stretchesout the light that reaches us and extends itswavelength towards the red end of the spec-trum. The first stars are thought to have startedshining less than a billion years after the BigBang, giving them ‘redshifts’ in which thechange in the light’s wavelength relative to itsoriginal value is 20 or more — moving visiblewavelengths well into the infrared. This is alarge part of the reason why even far-sightedHubble has never seen objects with a redshift ofmore than 7. The Webb telescope should solvethis: young stars characteristically give offultraviolet light that, after a redshift of 15,shines at 1.9 �m — “smack in the middle of ourbest band”, enthuses John Mather, the JWSTsenior project scientist.

First lightRecently, Mather was part of a team led byAlexander Kashlinsky, now of Goddard SpaceFlight Center near Washington DC, that usedthe much smaller Spitzer infrared space tele-scope to detect a diffuse glow from ‘first light’stars (A. Kashlinsky et al. Nature 438, 45–50;2005). No current or planned telescope, noteven the Webb telescope, can resolve individ-ual first-light stars. But the Webb telescopeshould be able to see the supernovae thatresulted when these massive but short-livedbodies exploded, providing the Universe withits first heavy elements.

It should also see the first galaxies thatformed. One of the key observations for theWebb telescope will be ‘deep field’ picturessimilar to those taken by Hubble. In these, atelescope points at a small patch of sky, takinga long, deep exposure that is designed to revealextremely faint, distant objects. Astronomershope the Webb telescope’s near-infrared deepfield (where contrast and resolution are best)will provide them with images of the very firstgalaxies and proto-galaxies.

For all this, advocates of the Webb telescopeare eager to point out that it is more than just a‘first light’ machine. They argue — especiallyto scientists whose projects are being sacri-ficed — that although the Webb telescope wasinspired by cosmologists’ interest in the earli-est stars, it has much to offer other fields.

Take, for example, the search for planetsaround other stars. One of the casualties ofthis year’s NASA budget was the TerrestrialPlanet Finder, a mission designed to look forobjects the size of Earth; its budget fell to zero.The Webb telescope cannot do what the Ter-restrial Planet Finder was meant to do. But

planetary scientist Jonathan Lunine of theUniversity of Arizona, Tucson, points out thatit should still deliver relevant science that nocurrent telescope can. An interdisciplinaryinvestigator on the telescope’s science workinggroup, Lunine says it will return images andspectra for planets not all that much biggerthan Jupiter, and may in special circumstancesproduce spectra for the atmospheres of plan-ets as small as Uranus. Its high-resolution pic-tures of dusty circumstellar disks will be thesharpest ever, providing insight into planetformation. It even has applications within ourown Solar System, for studying the thermalproperties of the Kuiper-belt objects that orbitbeyond Neptune.

And these are just the planned observations.Heidi Hammel, a planetary astronomer at theSpace Science Institute in Boulder, Colorado,and another member of the JWST scienceworking group, says some of the telescope’smost important results may well be unfore-seen. Some of Hubble’s best findings, includingthe deep-field observations, “came from things

we hadn’t even thought of, because it openedup new discovery space”, she says.

Cash registerSo no one is denying that the JWST will be afirst-rate telescope, perhaps even a revolution-ary one. Just last August an independentassessment team charged by the project toreview the telescope’s science potentialreported that “the scientific case for the JWSTmission has become even stronger” since theDecadal Survey’s endorsement in 2000. Butwhat of its expense? NASA’s latest budget putsthe project’s price tag, including $1 billion fora decade’s worth of operations, at $4.5 billion.That’s more than the entire annual researchand development budget of the National Sci-ence Foundation; it represents more than$1 million for each full member of the Ameri-can Astronomical Society.

On top of that there are the contributions byEurope and Canada, junior partners on thistelescope just as they were on Hubble. Europewill contribute one of the telescope’s four

Sun screen: this half-scale model shows how the Webb telescope will be shielded from solar heat.

NO

RTH

ROP

GRU

MM

AN




NEWS FEATURE NATURE|Vol 440|9 March 2006

142

1.7 �m, on the basis that future large ground-based telescopes equipped with adaptive opticswould be able to deal with these wavelengthsmore-or-less as well. And switching from a vac-

uum test chamber inOhio to one at theJohnson Space Centerin Houston, Texas,should save more than$100 million.

Despite this, thecost has continued to climb, alarmingly jump-ing almost $1 billion in 2005 alone (see graph).NASA’s requirement that the pro-gramme beef up its contingency fundadded a little over $200 million. Adelay in the government’s decisionto move from a US launcher tothe Ariane added an estimated$300 million as highly paidengineers were unable tomove forward until theyknew which rocket theywere designing for.The situation is particularly em-barrassing giventhat the cost of delaying thedecision endedup being greaterthan the cost of thelaunch. That delay,and a NASA decisionto rearrange the project’slong-term budget yetagain, saw the launch slipfrom 2011 to its current dateof 2013. Every slip increases thetotal cost.

By the standards of ground-basedastronomy, just a year’s worth of Webbtelescope overrun looks vast. Even themost expensive proposed instruments,such as the Atacama Large Millimeter Array(see Nature 439, 526–528; 2006) or the variousKeck-dwarfing 30-metre telescopes that areunder discussion, should leave ample change

from $1 billion. But O’Delloffers some perspective. Spacetelescopes are more expensivenot just because the technologyis more challenging, butbecause every problem andevery contingency has to bethought through and solvedbefore launch. This typicallyrequires a large team of engi-neers to remain in place foryears. What seem to be addi-tional costs have also comefrom NASA’s long and painfulswitch to ‘full-cost accounting’.In this system, all of a mission’sexpenses — every paper clipand every guard at the frontgate — are included in the total

bill. This makes NASA overheads smaller, andthe prices of individual missions greater.

For all this, the growth in cost of the Webbtelescope is not unprecedented. O’Dell recallsthat in 1972, Hubble’s total price including itsfirst year of operation was projected to be$300 million ($1 billion in today’s prices).According to Robert Smith, a historian atCanada’s University of Alberta who wrote apolitical history of the telescope, Hubbleended up costing a lot more by the time itreached the launch pad in 1990 (severalyears late owing to the Challenger shuttle

accident). He says that if the budgets were cal-culated according to NASA’s current full-costaccounting standards, “the development costof Hubble to date is certainly more than $4 bil-lion in today’s dollars”.

NASA’s Eric Smith adds that when newinstruments and operating expenses areadded, that comes to $9 billion. This doesn’tinclude the cost of four space-shuttle servicingmissions to Hubble, and a fifth being planned— the cost of a shuttle launch can be put atabout $500 million. All in all, building, launch-ing, using and refurbishing Hubble has proba-bly been the most expensive undertaking evermade in the name of pure science; the missionis still, remarkably, costing more than$300 million a year.

In that context, you begin to understandhow Lunine can claim with a straight face thatthe Webb telescope — which will outperformHubble in almost every way — is in fact “a bargain” at $4.5 billion. Still, the discipline as awhole has to wonder whether it can afford a

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

PROJECTED COST OF THE JAMES WEBB SPACE TELESCOPE

1996 1998 2000 2002 2004 20060

4,500

US$

mill

ion

in 2

00

6 do

llars

Hubble Space Telescope*

Price of design and development, with full-cost accounting*

instruments, the Near Infrared Spectrograph,and the launch on an Ariane 5 rocket. For aninvestment that approaches half a billion dol-lars, it will get 15% of the viewing time.Canada’s $57 millionwill provide a fineguidance sensor andother hardware, forwhich it gets about5% of the science use.

The total cost ismore than 30 times greater than that of theKeck telescopes on Hawaii, which boast two ofthe largest mirrors on Earth. One reason forthis extraordinary expense is that the JWST is achallenging spacecraft to build. The segmentedstructure of the mirror, made from 18 hexago-nal pieces of beryllium, is unlike anything builtbefore; so is the multilayer sunshade and thesystem that will deploy them both. RobertO’Dell, who as project scientist for Hubble wasin Mather’s position 30 years ago, points outthat Hubble was able to borrow much of itstechnology from spy satellites. The Webb tele-scope has no such heritage on which to draw.

Astronomical costsNASA tried to head off difficulties by tacklingsome key technology issues early in the pro-ject’s life. Although that helped to identifypotential trouble spots, it didn’t always reducecosts, says Eric Smith, programme scientist forthe Webb telescope at NASA headquarters inWashington DC. For example, the engineersfound they could build lightweight mirror seg-ments, but not as fast as some had hoped —the job will end up taking six years instead offour. The early development work led to themirror losing a third of its originally envisagedsurface area in 2001. Other proposed cuts incapability — the dropping of the telescope’smid-infrared instrument, a possible furthershrinkage to the mirror — were deemed scien-tifically unacceptable.

Some savings have been found. In 2005, pro-ject managers decided to forgo the extra mirrorpolishing needed to make the telescope’s imagesutterly crisp in wavelengths shorter than

“It’s like rush hour traffic: suddenlyeverybody’s piled up, there are

delays, and it becomes unstable.” — David Black

SOU

RCE:

E. P

. SM

ITH

, NA

SA




143

guys are so stupid. Why can’t you do this forless?’ That’s now haunting NASA, of course.”

Today, Illingworth inveighs against the“extraordinarily bad, artificial cost estimates”of the Goldin era. But the 2000 Decadal Surveyseems to have been happy to accept them. Theworld of big science is well used to projectsbeing lowballed — a process that gets schemesstarted on the basis of a low cost estimate, withthe implicit hope that by the time the truecosts are known inertia and vested interestswill make it impossible to pull out. Lowballingis not a practice anyone would defend on prin-ciple, but histories like the Hubble’s show it canwork (see page 127).

Past the disquietCraig Wheeler, a University of Texas astro-physicist and president-elect of the AmericanAstronomical Society, takes the lessons of Hub-ble to heart: “I remember when we were build-ing the Hubble Space Telescope, which has beenspectacularly successful, there were an awful lotof eggs put in that basket. And other smaller,faster, university-based projects suffered. Ithink we got through it.” Wheeler accepts thedisquiet over the Webb telescope’s costs, but he doesn’t think astronomers have yet reached“the point we collectively would say ‘enough’”.

And he warns againstrevisiting the results ofthe Decadal Survey onthe basis of the currentcrisis: “You alter thosepriorities at great risk.”

Kulkarni is morepessimistic. He thinks that NASA’s “laserlikefocus” on the Webb telescope short-changesmissions that would hunt for planets, probethe nature of dark matter, search for gravita-tional waves, and tackle other topics thatmight ultimately prove more popular withyoung scientists and with the public.

His concern is shared by Charles Beichmanof the Jet Propulsion Laboratory in Pasadena,a leading light of the cancelled TerrestrialPlanet Finder mission. Beichman thinks theWebb telescope will be “a fine machine. It willdo fantastic science”. In fact, he is on one of the instrument teams. But when he goes toprofessional meetings, he sees more youngastronomers attending sessions on planet-finding than on Hubble or the Webb telescope.

Lunine thinks the critics are fighting thewrong battle (and that anyone who doesn’trealize that the Terrestrial Planet Finderwould be costlier than the Webb telescope isdreaming). It is not the JWST that is to blame,he says. The real problem is that “NASA’s sci-ence budget is not adequate”, and science is“taking the hit” as the agency shifts its focus toreturning astronauts to the Moon. That maybe the case. But for now, the Webb telescope isleft in the awkward position of being the onlyone eating in a room full of hungry people. ■

Tony Reichhardt writes for Nature fromWashington DC.

bargain quite this big in straitened times.Between them, Hubble and the Webb tele-scope will soon consume half of NASA’s astro-physics budget (see chart).

Some critics have concluded that the care-fully crafted recommendations of the mostrecent Decadal Survey are no longer viable: ifthe costs had been clear, the priorities mightwell have been different. Kulkarni was on thereview panel for ultraviolet, optical andinfrared astronomy from space. He says, “Inow regret that we were not clear thinkers”about what was affordable, and he believes thatthe plan was “fiscally unrealistic” even beforeNASA cut its science budget last month.

If it was unrealistic, says David Black, presi-dent of the Universities Space Research Asso-ciation, scientists should share in the blame.“Astronomers pushed NASA to have all thesemissions. NASA bought into that. And all it

takes is one hiccough like the JWST overrun.It’s like rush hour traffic. One incident, and

suddenly everybody’s piled up, there areschedule delays, and it becomes unstable

very quickly.” The problems that come with send-

ing mission after mission into thiscrowded traffic are exacerbated

when the costs of the missionsare set artificially low at the

beginning. When NASAadministrator Mike Griffin

told a January meeting of the American Astro-nomical Society that the

Webb telescope wasn’t so much overbudget today

as it was “undercosted” at its inception, he wasn’t just

putting a good spin on things.The Decadal Survey guessed the

cost as $1 billion. Studies in the mid-1990s had pegged the price at between

$500 million and $1 billion. These werebased partly on the hope — unfulfilled, as

it happened — that the Webb telescopemight take advantage of advances in building

low-cost spacecraft developed by the military.Oddly, earlier cost estimates for a largeinfrared space tele-scope were closer tothe mark. A 1984Space Science Boardpanel predicted thecost including opera-tions to be $4 billion(roughly $7 billion today), and a subpanel ofthe 1990 Decadal Survey thought it would runto about $2 billion not counting operations,which, when adjusted for inflation, closelymatches NASA’s current projections.

Political playsGarth Illingworth of the University of Califor-nia, Santa Cruz, who chaired the 1990 panel,chalks the anomalously low estimates from the1990s up to a “lack of reality” inherent in the‘faster, better, cheaper’ philosophy of DanGoldin, NASA’s administrator at the time.Goldin focused on accelerating the develop-ment of spacecraft, and increasing innovation,while accepting a moderate rise in the risk offailure. Some projects conceived under thistag, in particular two Mars missions lost inquick succession, brought it a certain disre-pute. “It was a horrible, political circumstanceframing all the discussion in that decade,” saysIllingworth.

Reinhard Genzel of Germany’s Max PlanckInstitute for Extraterrestrial Physics in Garch-ing says it was clear at the time that a $500-mil-lion estimate for the Webb telescope was a“political price”. Yet such was the climate of the1990s that when estimates for the EuropeanSpace Agency’s smaller Herschel infraredobservatory came in at $1 billion, he says, “I wasapproached by many colleagues saying, ‘You

Far out: the Webb telescope will have an array of hexagonal mirrors (left) that work to find themost distant galaxies yet.

NATURE|Vol 440|9 March 2006 NEWS FEATURE

NASA’S ASTROPHYSICS BUDGET US$1.5 billion in 2008

JWST

Hubble

BALL

AER

OSP

AC

E &

TEC

HN

OLO

GIE

S ; N

ASA

““I remember when we werebuilding Hubble, other projects

suffered. I think we got through it.”— Craig Wheeler

SOU

RCE:

NA

SA



Documents

US astronomy: Is the next big thing too big?