Measuring the Hubble constantMario Livio and Adam G. Riess Citation: Phys. Today 66(10), 41 (2013); doi: 10.1063/PT.3.2148 View online: http://dx.doi.org/10.1063/PT.3.2148 View Table of Contents: http://www.physicstoday.org/resource/1/PHTOAD/v66/i10 Published by the AIP Publishing LLC. Additional resources for Physics TodayHomepage: http://www.physicstoday.org/ Information: http://www.physicstoday.org/about_us Daily Edition: http://www.physicstoday.org/daily_edition

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

In 2011 a passionate debate flared up about whodeserves the credit for the discovery that our uni-verse is expanding. Here are some of the back-ground facts. By February 1922, US astronomerVesto Slipher had already measured the redshifts

for 41 galaxies. British astrophysicist Arthur Ed-dington, who listed them in his 1923 book The Math-ematical Theory of Relativity (Cambridge UniversityPress), noted that “the great preponderance of positive (receding) velocities is very striking.” Hedid add, however, that the lack of observations fromthe Southern Hemisphere precluded any definitiveconclusions.

In 1924 Swedish astronomer Knut Lundmarkprovided tentative, qualitative evidence for the ex-pansion. However, his results did not carry muchweight, since he relied on the implausible assump-tion that all galaxies have the same diameter and hiscorrelation between velocity and distance was notreadily apparent.

A stronger case for an expanding universecame from Belgian priest and cosmologist GeorgesLemaître who, in 1927, published a paper in Frenchentitled “A homogeneous universe of constant massand increasing radius accounting for the radial velocity of extragalactic nebulae.” In that paper,Lemaître reported on the expanding-universe solu-tions to Einstein’s general relativity equations. Healso used Slipher’s results in combination with dis-tance estimates, now based on the rather inaccurateassumption that all galaxies have the same luminos-ity, to propose a tentative “Hubble law” v = H0d, thatis, a linear relationship between distance d and re-cession velocity v. Lemaître derived the value of625 kilometers per second per megaparsec for the

Hubble constant H0. (The expansion rate actuallychanges with time as 1/t; the “constant” H0 is itspresent value.)

Unfortunately, Lemaître’s paper received littleattention. Two years later Edwin Hubble, shown infigure 1, published his seminal paper with im-proved distance determinations based on thebrightnesses of certain classes of stars. His resultinglinear relation between recession velocity—again,obtained from Slipher’s redshift data—and distancewas more significant and convincing. It became thewidely cited origin of the discovery of the expand-ing universe. The value Hubble obtained for theHubble constant was 500 km s−1 Mpc−1.

A key driver for the recent debate over who dis-covered the Hubble law was that in the Englishtranslation of Lemaître’s paper, which appeared in1931, certain paragraphs were omitted. Suspicionarose that the omission reflected some form of cen-sorship, possibly encouraged by Hubble. After ex-tensive research, one of us (Livio) discovered con-clusive evidence that, following the publication ofHubble’s more precise results, Lemaître, out ofmodesty, deleted those paragraphs from the trans-lation because he thought they were superseded byHubble’s work.1 In brief, Lundmark was the first tooffer observational evidence for the expansion,Lemaître made the tentative connection betweentheory and observations, Hubble and his assistant

www.physicstoday.org October 2013 Physics Today 41

Mario Livio is an astrophysicist at the Space Telescope Science Institute in Baltimore, Maryland. Adam Riess is an astrophysicist at the STScI and a professor of astronomy and physics at the Johns Hopkins University in Baltimore.

Measuring the HUBBLE CONSTANT

Mario Livio and Adam G. Riess

Perhaps the fundamental parameterof cosmology, the ratio of an object’srecessional speed to its distancefrom us encodes information aboutthe universe’s age, composition, and structure.

The Hubble Space Telescope orbitingEarth in 2009. (Courtesy of the Space Telescope Science Institute and NASA.)

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Milton Humason provided the best observationalproof in a series of papers, and all were aided bySlipher’s redshifts.

Early measurementsEver since the 1920s, physicists have known that welive in an expanding universe. In the framework ofgeneral relativity and given the “cosmological prin-ciple” that the universe is the same at all locationsand in all directions on large scales, the expansionis measured by a single function of time, the scalefactor R(t). The evolution of the scale factor is gov-erned by what is known as the Friedmann equation,which describes how the curvature of space is de-termined by the density of matter and radiation andthe density of the so-called dark energy, the physicalmanifestation of Einstein’s cosmological constant.The most recent observations of the cosmic micro -wave background, by the Wilkinson MicrowaveAnisotropy Probe (WMAP) and the Planck space ob-servatory, indicate that matter and dark energycombine so as to produce nearly, if not exactly, zerospatial curvature.

The Hubble constant is a direct measure of thecurrent expansion rate and is the key parameter indetermining the age of the universe, t0, throught0 ∝ H0

−1. (Matter and dark energy play a smaller role,causing the cosmic expansion to decelerate or accel-erate, respectively.) Physical processes such as thegrowth of cosmic structure and the nucleo synthesisof the light isotopes hydrogen, deuterium, helium-3, helium-4, and lithium-7 depend on the cosmic ex-pansion rate and thereby on the value of H0. So docritical periods in the universe’s history, such as thetransition from a radiation-dominated epoch inwhich the energy density scales as R−4(t) to the matter- dominated universe in which density is proportional to R−3(t). It should therefore come

as no surprise that the determination of H0 has been a major observational goal for the past ninedecades.

Figure 2a summarizes the early history of H0measurements. The first values were all roughly500 km s−1 Mpc−1, with an uncertainty estimatedrather naively “to be of the order of 10 percent.”2

Some 20 years after Hubble’s measurements, as-tronomer Walter Baade revised the distance tonearby galaxies. Baade recognized that Hubble hadconfused two generations of pulsating stars used asdistance standards; his revised estimates increaseddistances and reduced H0 by about a factor of two.Measurements of H0 first approached the range ofvalues accepted today (see figure 2b for modernmeasurements) through the work of Allan Sandagein the late 1950s. Sandage demonstrated that Hub-ble had mistakenly identified ionized-hydrogen(H II) regions as bright stars and revised the valuedown to H0 ≈ 75 km s−1 Mpc−1, recognizing that theuncertainty was comparable to the value itself.

In the three decades that followed, publishedvalues of the Hubble constant ranged over a factorof two. One group, led by Sandage, consistentlyclaimed values near 50 km s−1 Mpc−1 ± 10%; another,led by Gérard de Vaucouleurs, persistently meas-ured 100 km s−1 Mpc−1 ± 10%. Those determinationswere embarrassing for two reasons. One, obviously,was the smallness of the cited errors, which couldn’tboth be correct. The other was that if one took the average of the two values, the deduced age ofthe universe was shorter than the calculated ages of the oldest star clusters! The eventual resolution ofthe age discrepancy required two inputs: accurateastrometric distance determinations by the Hipparcossatellite, which somewhat reduced the calculatedstellar ages, and the discovery that the cosmic ex-pansion is accelerating, which implied an older agefor the universe.

To determine redshifts, and therefore radial ve-locities, astrophysicists need to look far enoughaway that cosmic expansion rather than local mo-tions predominantly determine the redshift. Giventhat, and notwithstanding the challenges posed bythe most distant or faintest objects, redshifts can beobtained relatively readily; the problem of meas -uring the Hubble constant has always primarilybeen one of determining accurate astronomical distances. The availability of new observing facili-ties—the Hubble Space Telescope in particular—hasallowed for a dramatic improvement in distance determinations.

Cepheids, a primary distance indicatorThe simplest and most robust measure of distancecomes from geometry via triangulation, with the di-ameter of Earth’s orbit around the Sun serving as abaseline. Unfortunately, the parallax essential fortriangulation can only be detected for stars withinthe Milky Way at distances of up to about 1 kilo -parsec; for even the nearest galaxies, the angular shiftwould be less than a microarcsecond. Consequently,astrophysicists must use other methods to deter-mine extragalactic distances of tens to hundreds ofmegaparsecs.

42 October 2013 Physics Today www.physicstoday.org

Hubble constant

Figure 1. Edwin Hubble with an image of a galaxy. Hubble presentedconvincing evidence that galaxies recede with a speed proportional totheir distance from us. (Hale Observatories, courtesy of the AIP EmilioSegrè Visual Archives.)

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

The distances so obtained are relative meas-ures, calibrated from nearby examples within rangeof parallax analysis, and the objects to which themethods are applied are called primary distance in-dicators. Secondary distance indicators are objectsor systems that are far away from us. Their intrinsicrarity places the nearest examples beyond the rangeof parallax, and so their distances are calibrated fromprimary indicators. The sequence continues, forminga distance ladder that reaches areas of the universewhere the smooth expansion of the cosmos—theHubble flow—dominates over local, peculiar mo-tions. Errors in measurement or systematics neces-sarily propagate along the ladder to the determina-tion of H0.

The most common relative distance indicatorsemploy so-called standard candles that are based ongeometrical properties, physical properties, or var-ious correlations. Underlying the notion of standardcandles is that the flux of radiation decreases as aninverse square law. Useful standard-candle candi-dates either have a constant luminosity or are ob-jects whose luminosity can be related to a measura-ble property that is independent of distance, such asan oscillation period or the decay rate of a transientlight curve. They are highly luminous and so can beseen far away.

Perhaps the best-known standard candles arepulsing, supergiant stars called Cepheid variables.They are named after the prototype of this class,Delta Cephei, which has a 5.4-day period and, at a distance of 270 parsecs, a parallax angle of3.7 milliarcseconds. The relative proximity to theSun of some Cepheids in the Milky Way, their highluminosity (104–105 times the Sun’s), and well- understood stellar physics make them one of the

most reliable of the primary distance indicators. In1912 Henrietta Leavitt, shown in figure 3, realizedthat the relation between the Cepheids’ period andluminosity could allow those giant stars to serve as standard candles. Hubble used the Cepheids todetermine distances to Local Group galaxies, thegroup of about three dozen nearby galaxies towhich the Milky Way belongs.

The physical process responsible for aCepheid’s pulsations and the period–luminosity re-lation begins with a perturbation—a contractionthat drives the star away from hydrostatic equilib-rium. As a result, the temperature and density of thegas increase. Near the star’s ionization zones—inwhich the primary reaction is He+ � He++—as he-lium becomes doubly ionized, the opacity of the gasincreases. Consequently, the gas traps radiation,warms up even more, and further expands. In theionization zones, the pressure does not decrease asquickly with decreasing density as it would for anideal gas, so the expansion goes beyond the pointwhere hydrostatic equilibrium would have beenachieved for an ideal gas. The excess heating stopsonly when helium recombination back into He+ re-duces the gas opacity and heat is radiated intospace. As heat leaves the system, the pressure de-creases, the gas contracts, and the cycle starts anew.The pulsation cycle operates only for a small tem-perature range, and only in a narrow range of stellarmasses does the ionization zone occur at a middlingdepth where ionization and recombination can ef-fectively drive oscillations.

It turns out that the pulsation period is inverselyproportional to the square root of the Cepheid’sdensity and therefore depends on the stellar massand radius. The luminosity is determined by some

www.physicstoday.org October 2013 Physics Today 43

YEAR

H0

(km

sM

pc

)−1

−1

800

600

0

200

400

1920

700

500

300

100

1930 1940 1950 1960 1970 1980

Lemaître

Hubble and Humason

Hubble

Mineur

Behr

Baade and Thackeray

McVittie

Hubble

Humason, Mayall, and Sandage

Sandage

Ambartsumyan Sandage and Tammann

Holmbergvan den Bergh

de VaucouleursSérsic

a

YEAR

H0

(km

sM

pc

)−1

−1

60

01990

20

40

80

100

120

1995 2000 2005 2010

b

Figure 2. Hubble constant

measurements. (a) The earliest measurements of H0 were about anorder of magnitude greater thanmeasurements made 50 years later.(Adapted from ref. 10.) (b) Moremodern measurements have tended to cluster around a range of 70–75 km s−1 Mpc−1, particularlysince 2005 (red band). The most re-cent and precise measurement7 hasan uncertainty of 3%; with improved instrumentation and data handling,astronomers hope to soon achieve aprecision of 1%.

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

power of the mass, but it is also a simple function oftemperature and radius. Given the three specific formulas, one expects (and astronomers have ob-served) a temperature-dependent period–luminosityrelation.

Thanks to the high optical resolution and sta-bility of the Hubble Space Telescope and the capabili-ties of the Wide Field Planetary Camera 2 (WFPC2)instrument installed at the first Hubble servicingmission, observers could resolve individual Cepheidsin galaxies as far away as 20 Mpc. One of Hubble’sinitial “key projects,” led by Wendy Freedman, Jeremy Mould, and Robert Kennicutt, and anotherinvestigation led by Sandage, Gustav Tammann,and Abi Saha, took advantage of that then-new ca-pability to measure the Hubble constant with about10% precision. The key project, which operatedfrom 1994 to 2001, and several contemporaneousprograms also used Cepheids to calibrate tools foranalyzing many of the secondary distance indica-tors mentioned later.

Below we briefly describe the physical basis forlong-range distance methods that are presentlybeing used to refine determinations of the Hubbleconstant. Of the three techniques most commonlyemployed, one—analysis of supernova lightcurves—rests on calibration via Cepheids. But ap-proaches based on so-called megamasers and grav-itational lensing are at least partially geometric andthus serve to indicate distances independently ofCepheid data.

Extending the ladderType Ia supernovae result from the thermonucleardisruptions of mass-accreting white dwarfs madeup of carbon and oxygen. At peak brightness, theyare extremely bright—a million times brighter thanthe Cepheids—and, as a class, they show relativelylittle variation in that peak luminosity. Furthermore,there exists a tight correlation between the peak lu-minosity of a type Ia supernova and the shape of its

brightness-versus-time light curve. As figure 4shows, the light curve declines more slowly forbrighter supernovae. (See also the article by SaulPerlmutter, PHYSICS TODAY, April 2003, page 53.)

Physicists currently lack a full theoretical un-derstanding of the observed correlations, but therelative homogeneity likely arises because the whitedwarfs that blow up as supernovae are at or near theChandrasekhar limit, the mass above which gravityovercomes electron degeneracy pressure and run-away nuclear fusion ensues. The relation betweenpeak luminosity and light-curve shape may be theresult of the following scenario. The peak luminos-ity of a type Ia supernova is proportional to the massof nickel-56 that it produces, since the decay of 56Niis what powers the luminosity. A higher mass of56Ni, however, also results in more heating and ahigher opacity; consequently, the light curve de-clines more slowly. On its way to us, light from supernovae passes through dust, which reduces itsintensity and shifts its color to the red.

Once the light-curve and dust-induced varia-tions are accounted for, type Ia supernovae have astatistical dispersion of about 15% in their optical lu-minosity and about 10% in the near-IR. The low dis-persion and high luminosity of type Ia supernovaehave made them the most widely sought long-rangestandard candle. They can be used to measure dis-tances over a range of 40–5000 Mpc, but they arerare; about one supernova goes off each decadewithin a distance of about 20 Mpc from us. Thusthere are only a few good examples near enoughthat their host galaxies also include visibleCepheids, and some of those supernovae date to theera of less accurate photographic observations.

The megamaser method,3 first applied to thegalaxy NGC 4258 (shown in figure 5), some 8 Mpcaway, has proved to be an effective approach formaking direct distance measurements beyond theLocal Group. The technique involves submilliarc-second-resolution imaging and monitoring ofwater-vapor maser emission from circum nucleardisks surrounding the black hole in active galaxiessuch as NGC 4258. Basically, if a masing blob that isacted on mainly by the gravitational pull of the cen-tral black hole moves in the disk with a velocity vand centripetal acceleration a, then the distance tothe disk can be determined through

where ΔΘ is the apparent angular radius of the orbitand i is the orbit’s inclination, with i = 90° denotingthat the orbit is seen edge on.

The inclination can be determined via the an-gular offset of “systemic” masers from the centralblack hole. (Systemic masers have velocities close tothe recession velocity of the galaxy.) The other pa-rameters on the right-hand side of the equation canbe determined with very long baseline interferom-etry observations of blueshifted and redshiftedblobs coupled with monitoring of maser spectraover a long period of time. Thus the distance can beaccurately determined. Unfortunately, astronomershave not found any other megamaser system that

D i= sin ,v2

aΔΘ

44 October 2013 Physics Today www.physicstoday.org

Hubble constant

Figure 3. Henrietta Leavitt, seated at her desk. Leavitt recognized thatthe relationship between the period and the luminosity of supergiantstars called Cepheid variables implied that the Cepheids would be gooddistance standards. (Courtesy of the AIP Emilio Segrè Visual Archives,PHYSICS TODAY Collection.)

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

can be observed with a precision comparable to thatfor NGC 4258; indeed, the potential precision of thenext-best observed candidates—less than a dozen—is several times worse. Rarity of megamaser systemsmay thus limit the ultimate precision achievable bythe method.

When a galaxy lies along the line of sight to amore distant source galaxy, strong gravitationallensing can produce multiple images of the source.If the source happens to be variable, the time delaysbetween the different images can be accuratelymeasured through careful monitoring of lightcurves. When combined with a model for the massdistribution in the gravitational lens, which deter-mines the gradient of the gravitational potential, thetime delays can be used to convert angular separa-tions into absolute distance. As with megamasers,rarity of suitable systems may ultimately be a sig-nificant limitation of the lensing method.

A miscellany of methodsOther distance indicators have contributed less inrecent times to the accurate measurement of theHubble constant due to their larger systematic errors or complexity. Those include the Tully–Fisherrelation between the luminosity and maximum ro-tational velocity for a spiral galaxy; the fundamentalplane, a correlation among the effective radius, ef-fective surface brightness, and central velocity dis-persion in elliptical galaxies; the tip of the red-giantbranch, a method based on the luminosity of thebrightest red giants in a galaxy; and methods rely-ing on core- collapse supernovae, globular clusters,or planetary nebulae.

Two rather interesting methods have, unfortu-nately, failed thus far to produce very accurate re-sults. One is based on the Sunyaev–Zeldovich effect.The phenomenon involves the hot (kT ≈ 10 keV) gastrapped in the potential wells of galaxy clusters. Pho-tons from the cosmic microwave background thatpass through a cluster have a small probability of in-teracting with energetic electrons in the intracluster

gas. The inverse Compton scattering that ensuesboosts the energy of the microwave backgroundphoton and, in turn, generates a small distortion inthe spectrum of the microwave background at fre-quencies near 218 GHz—slightly more photons athigher frequencies and slightly fewer at lower fre-quencies than would otherwise be expected.

The magnitude of the effect is proportional tothe integral of the pressure along the line of sight,∫neTedl, where ne and Te are the electron number den-sity and temperature, respectively. Since the x-rayemission from the intracluster medium is propor-tional to a different power of the density, a combi-nation of the two measurements can, given some as-sumptions about the cluster geometry, determinethe distance to the cluster without the need to cali-brate against other standard candles. At present,systematic uncertainties associated with the methodare 15–30%.

The second physical method is that of surfacebrightness fluctuations. That method, developedprimarily by astronomer John Tonry and collabora-tors, exploits the obvious fact that the ability to re-solve stars within galaxies is distance dependent.4

More specifically, for every region of a galaxy onecan measure the average flux per pixel, g, and the pixel-to- pixel root-mean-square variation, σ. Sincethe flux obtained in a pixel is received from N starsof average flux f‾, g = N f‾ and σ = √N‾‾ f‾. A galaxy thatis twice as distant thus appears twice as smooth asthe closer galaxy, because N is proportional to dis-tance squared. Although elegant, the method israther difficult to employ because it requires the ex-cision of globular clusters and background galaxies,which produce large fluctuations, and of areas ofdust absorption, which smooth fluctuations.

The Hubble constant and dark energyThe new standard model of cosmology, largely es-tablished in the past 15 years, is dominated by twopoorly understood components, dark matter anddark energy. Dark energy, the cause attributed to the

www.physicstoday.org October 2013 Physics Today 45

−20 −20

−19 −19

−18 −18

−17 −17

−16 −16

−15 −15

AB

SO

LU

TE

MA

GN

ITU

DE

SC

AL

ED

MA

GN

ITU

DE

−20−20 00 2020 4040 6060

SCALED DAYSDAYS

a b

Figure 4. Type Ia supernova light curves. (a) As shown in these plots of magnitude (a logarithmic measure of luminosity;more negative values are brighter) versus time, luminosity decays more slowly for the brightest supernovae than it does for the fainter ones. (b) With a suitable rescaling of time and brightness, the supernovae plotted in panel a lie on a single curve. Because of that universality, type Ia supernovae can serve as standards for determining distance. (Courtesy of the SupernovaCosmology Project.)

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

accelerating expansion of theuniverse, is the larger enigma.It may represent the energy ofthe vacuum, but in that case itsobserved density is puzzlinglylow. It may be associated withsome scalar field, but the timedependence of that field re-mains an open question.

The empirical approach tounderstanding dark energy isto measure its equation-of-stateparameter, w = P/ρc2, where Pis the pressure, ρ is the density,and c is the speed of light. If thedark energy represents vac-uum energy (or, equivalently,Einstein’s cosmological con-stant), then w = −1. Alterna-tively, the cosmic accelerationcould result from a decayingscalar field, a low- energy imita-tion of the field that drives in-flation. If so, it may be possibleto detect a time dependence ofw or a scale dependence indicating a breakdown ofgeneral relativity.

A wide range of ultraprecise cosmologicalmeasurements in the planning stages may addressthe nature of the dark energy. Measurements of theHubble constant based on local objects with red-shifts of up to about 0.1 should yield a precision ap-proaching 1%. Combined with the precise cosmicmicrowave background observations from WMAP,the Atacama Cosmology Telescope, the South PoleTelescope, and the Planck satellite, those measure-ments can constrain w to about 2%.

Local measurements of H0 are complementaryto other, higher-redshift probes. Indeed, we’d be re-miss if we did not note an apparent tension, at the3σ level, between current measurements of H0 basedon local objects and its deduced value based on thestandard cosmological model and new Planck re-sults for the cosmic microwave background. Thattension may be the harbinger of new physics, butpast experience indicates that discrepancies below3σ often disappear when more data are available.

Toward more precise determinationsTo achieve the goal of 1% precision in local distancemeasurements requires refinements to mitigate thesystematic errors that dominated prior uncertain-ties. Here we focus on progress with the distance-ladder approach, though we expect progress onother fronts as well.

Only the better distance indicators are likely toyield improved precision. Moreover, they willachieve that end only if, to the extent possible, allthe individual exemplars are treated identically inconstructing the distance ladder. The availability ofmultiple types of distance indicators may offer away to crosscheck distance determinations, but onlyif the various indicators are independent and offercomparable precision.

With a number of crucial refinements made

possible in recent years, the ladder connecting geo-metrical distances to Cepheids and type Ia super-novae has the potential to reach the desired goal of1% precision. New instruments on Hubble—in par-ticular, the Advanced Camera for Surveys and theWide Field Camera 3 (WFC3)—have doubled thetelescope’s range and thus have afforded an eight-fold increase in opportunities for calibrating super-novae via Cepheids. Greater accuracy comes fromthe replacement of older, problematic data withmodern digital data. Past calibrations of type Ia su-pernovae via Cepheids made use of photographicplates obtained in 1937, 1960, and 1974. Those ana-log data could not readily benefit from modern digital techniques used, for example, to removebackground light from the host galaxy and ensurethat all observations are looking at the same portionof the spectral energy distribution. Other supernovadata obtained within the Wide Field Planetary Camera 2 range suffered from a missed peak in theluminosity or heavy extinction due to dust, or theywere atypical of type Ia supernovae. The SH0ES (Supernova H0 for the Equation of State) team led byone of us (Riess) and Lucas Macri is halfwaythrough calibrating 17 ideal type Ia supernovaewith the help of the new Hubble instruments.

After 10 years of radio observations of the watermegamasers in orbit around NGC 4258’s super -massive black hole, Elizabeth Humphreys and col-leagues have recently determined the distance to thegalaxy5 to be 7.60 Mpc, with an uncertainty of 3%.The improved distance estimate will serve to bettercalibrate Cepheids. New instruments on Hubble al-lowed Macri and colleagues to discover hundredsof Cepheids in NGC 4258 and to calibrate their luminosities.6 The SH0ES team observed thoseCepheids in the near-IR with the WFC3 to reducepast systematic uncertainties in the dependence ofCepheid luminosities on the star’s chemical compo-sition and on the variety of obscuring dust.

46 October 2013 Physics Today www.physicstoday.org

Hubble constant

Figure 5. The galaxy NGC 4258 contains many megamasers, radiatingmasses of gas whose speed and centripetal acceleration allow observers todetermine the distance to the host galaxy. (Courtesy of the Space TelescopeScience Institute, NASA, and the European Space Agency.)

Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms

Another improvement came from observingCepheids with a given range of periods both ingalaxies hosting supernovae and in NGC 4258, theanchor of the SH0ES distance ladder. Armed withsuch observations, astrophysicists can reduce sys-tematic uncertainties in the period–luminosity rela-tion. Observations of Cepheids in supernova hostgalaxies and NGC 4258 have now been taken withthe same instrument, a move that removes uncer-tainties in flux calibration. Those improvements reduced the uncertainty in the Hubble constant toabout 5% by 2009 and 3% two years later.7 Yet justas the SH0ES distance ladder was completed, con-struction had begun on a potentially more powerfuldistance ladder.

Trigonometric parallaxes to Cepheids in theMilky Way can, in principle, anchor a distance lad-der to reach 1% precision. Using the fine guidancesensors on board Hubble, in 2007 the astrometry sci-ence team led by G. Fritz Benedict measured theparallaxes of the 10 nearest Cepheids8 with a meanerror of 3%. Their sample provides an anchor for thedistance ladder that’s different from the one used bythe SH0ES team or the earlier key project. A ladderbuilt on that alternate anchor gives similar resultsfor the value of H0, but at the cost of giving back thereductions in systematic errors realized by theSH0ES team.

However, the new spatial-scanning capabilityof the WFC3 can provide even better parallax meas-urements and crucial flux measurements of brightCepheids. A group led by Riess and Stefano Caser-

tano has begun to use spatial scanning to measureparallaxes of the less common and previously un-calibrated longer-period Cepheids prevalent at dis-tances of 1–3 kpc. By the end of this decade the Eu-ropean Space Agency’s Gaia mission will alsoprovide Cepheid parallaxes out to 10 kpc. Recentmid-IR calibrations of Cepheids as part of theCarnegie Hubble Program can further improve thereliability of the local span of the distance ladder.9

If the measurements enabled by WFC3 are toserve as the anchor for a new, potent distance lad-der, the improvements in techniques related to sec-ondary distance indicators already achieved willalso need to be retained and extended. Type Ia su-pernovae have less scatter in the near-IR than theoptical, so near-IR measurements may put a 1%measurement of H0 within reach. When the new ladder is completed during Hubble’s third and likelyfinal decade, it will be a powerful tool, sturdyenough to probe the mysteries that still remain inthe standard cosmological model.

References1. M. Livio, Nature 479, 171 (2011).2. E. Hubble, M. L. Humason, Astrophys. J. 74, 43 (1931).3. J. Herrnstein et al., Nature 400, 539 (1999).4. J. L. Tonry, Astrophys. J. 373, L1 (1991).5. E. M. L. Humphreys et al., Astrophys. J. 775, 13 (2013).6. L. M. Macri et al., Astrophys. J. 652, 1133 (2006).7. A. G. Riess et al., Astrophys. J. 730, 119 (2011).8. G. F. Benedict et al., Astron. J. 133, 1810 (2007).9. A. J. Monson et al., Astrophys. J. 759, 146 (2012).

10. V. Trimble, http://arxiv.org/abs/1307.2289. ■

Some of over 100 Available FittingsUHV MULTI-CF FITTINGS

Customs Encouraged

www.kimballphysics.com email: info@kimphys.com

Tel. 603-878-1616Downloaded 01 Oct 2013 to 205.133.226.104. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://www.physicstoday.org/about_us/terms