Surveying the Accelerating Universe with Supernovae

Saurabh W. Jha Georgia Tech School of Physics Colloquium March 5, 2012

HST color composite of SN 2008A in NGC 634 from McCully et al. (in prep.) “Surveying” the Universe: Measuring Astronomical Distances the cosmological distance ladder

Leavitt (1912): 100th anniversary! Cepheid variable stars yield precise distances Hubble’s Law Hubble (1929)

measured distances using bright stars in other galaxies and average galaxy properties

measuredthe cosmological galaxy redshifts distance with ladder spectroscopy, using the Doppler shift of atomic spectral lines Hubble’s Law Hubble (1929)

wrong units!

measured distances using bright stars in other galaxies and average galaxy properties

measuredthe cosmological galaxy redshifts distance with ladder spectroscopy, using the Doppler shift of atomic spectral lines Hubble’s Law

theHubble’s cosmological Law is a consequence distance ladder of the expansion of the Universe v = H0 d History and Future of the Expansion History and Future of the Expansion

if the expansion were uniform, we could extrapolate backwards to get the age of the Universe: t0 = 1/H0 FLRW Cosmography general relativity + cosmological principle (homogeneity, isotropy) dr2 ds2 = c2 dt2 a(t)2 + r2 dθ2 + r2 sin2 θ dφ2 − 1 kr2 − a(t) scale factor k describes spatial curvature ≡ a λem 1 photons: ds2 =0 = = ⇒ a0 λobs 1+z FLRW Cosmography general relativity + cosmological principle (homogeneity, isotropy) dr2 ds2 = c2 dt2 a(t)2 + r2 dθ2 + r2 sin2 θ dφ2 − 1 kr2 − a(t) scale factor k describes spatial curvature ≡ a λem 1 photons: ds2 =0 = = ⇒ a0 λobs 1+z kinematic description of scale factor evolution a˙ aa¨ H q ... ≡ a ≡−a˙ 2 Taylor expand a(t) around t0: H 0, q 0, j 0, s 0, ... FLRW Cosmography general relativity + cosmological principle (homogeneity, isotropy) dr2 ds2 = c2 dt2 a(t)2 + r2 dθ2 + r2 sin2 θ dφ2 − 1 kr2 − a(t) scale factor k describes spatial curvature ≡ a λem 1 photons: ds2 =0 = = ⇒ a0 λobs 1+z kinematic description of scale factor evolution a˙ aa¨ H q ... ≡ a ≡−a˙ 2 Taylor expand a(t) around t0: H 0, q 0, j 0, s 0, ...

L f = 2 4πdL FLRW Cosmography general relativity + cosmological principle (homogeneity, isotropy) dr2 ds2 = c2 dt2 a(t)2 + r2 dθ2 + r2 sin2 θ dφ2 − 1 kr2 − a(t) scale factor k describes spatial curvature ≡ a λem 1 photons: ds2 =0 = = ⇒ a0 λobs 1+z kinematic description of scale factor evolution a˙ aa¨ H q ... ≡ a ≡−a˙ 2 Taylor expand a(t) around t0: H 0, q 0, j 0, s 0, ... intrinsic luminosity L f = 2 measuredﬂux 4πdL FLRW Cosmography general relativity + cosmological principle (homogeneity, isotropy) dr2 ds2 = c2 dt2 a(t)2 + r2 dθ2 + r2 sin2 θ dφ2 − 1 kr2 − a(t) scale factor k describes spatial curvature ≡ a λem 1 photons: ds2 =0 = = ⇒ a0 λobs 1+z kinematic description of scale factor evolution a˙ aa¨ H q ... ≡ a ≡−a˙ 2 Taylor expand a(t) around t0: H 0, q 0, j 0, s 0, ... intrinsic luminosity L z du f = dL(z)=c(1 + z) (k = 0) measured 4πd2 H(u) ﬂux L 0 History and Future of the Expansion a/a0 =1/(1 + z)

d /c ∼ L FLRW Cosmography connecting kinematics and dynamics a Universe full of perfect ﬂuids, with equations of state: P = w ρ c2 w equation of state parameter i i i ≡ 2 ρ (1 + z)3(1+wi) a˙ i ∝ ρi a¨ 2 a ∝ (ρi +3Pi/c )= ρi(1 + 3wi) a ∝− − FLRW Cosmography connecting kinematics and dynamics a Universe full of perfect ﬂuids, with equations of state: P = w ρ c2 w equation of state parameter i i i ≡ 2 ρ (1 + z)3(1+wi) a˙ i ∝ ρi a¨ 2 a ∝ (ρi +3Pi/c )= ρi(1 + 3wi) a ∝− − w density acceleration a(t)

matter 3 2/3 0 ρM (1 + z) ρM t (normal/dark) ∝ ∝− ∝ 4 1/2 radiation +⅓ ρ (1 + z) 2ρR t R ∝ ∝− ∝ cosmological Ht −1 ρΛ = const +2ρΛ e constant ∝ ∝ History and Future of the Expansion a/a0 =1/(1 + z)

different pasts and futures depend on gravity and the contents of the Universe

d /c ∼ L History and Future of the Expansion a/a0 =1/(1 + z)

different pasts and futures depend on gravity and the contents of the Universe

d /c ∼ L redshift yields the cosmic scale factor distance tells the time, but we need a good distance indicator Type Ia SN 1998bu in M96

March 14, 1997 Type Ia SN 1998bu in M96

March 14, 1997 May 18, 1998 Type Ia SN 1998bu in M96

one star in this galaxy becomes as bright as 10 billion stars!

March 14, 1997 May 18, 1998 SN 1998bu maximum light spectrum

intermediate mass elements are fused from C and O during the explosion ejecta travels at ~10,000 km/s

Fe Si Fe Ca S

Si ThermonuclearThe progenitor o(Typef a Type Ia)Ia s uSupernovaepernova

. . .which spills gas onto the Two normal stars The more massive secondary star, causing are in a binary pair. star becomes a giant. . . it to expand and become engulfed.

The secondary, lighter star The common envelope is and the core of the giant star ejected, while the separation The remaining core of spiral-in within a common between the core and the the giant collapses and envelope. secondary star decreases. becomes a white dwarf.

The aging companion The white dwarf's mass ...causing the companion star starts swelling, spilling increases until it reaches a star to be ejected away. gas onto the white dwarf. critical mass and explodes... the cosmological distance ladder a complicated and uncertain story, but we observe them to be very homogeneous Some “Nearby” SN Ia Light Curves

SN 1998bu in M 96 SN 2001V in NGC 3987 the groundbreaking work: Calán/Tololo SN Survey 1996AJ....112.2391H

shape of the light curve lets us read the label on our cosmic light bulb

measuring colors lets us correct for Phillips et al. (1993); Hamuy et al. (1996) attenuation of the light by dust SN Ia are calibrated candles:

luminosity correlated with light-curve decline rate (Phillips relation)

multicolor templates constrain dust

Krisciunas et al. (2006)

MLCS2k2 Jha, Riess, & Kirshner (2007) MLCS2k2 light curve ﬁts KAIT BVRI photometry Ganeshalingam et al. (2009, in prep)

μ = 33.46 ± 0.07 mag μ = 33.49 ± 0.10 mag MLCS2k2 light curve ﬁts KAIT BVRI photometry Ganeshalingam et al. (2009, in prep)

μ = 33.46 ± 0.07 mag μ = 33.49 ± 0.10 mag SN 1999cp and SN 2002cr, both in NGC 5468 Correcting for Intrinsic Variations and Dust

Si Correcting for Intrinsic Variations and Dust

empirical validation that the method works! Hubble’s Law extended

v = H0d

Si Hubble’s Law extended

v = H0d

our latest results from the HubbleSi Space Telescope: -1 -1 H0 = 73.8 ± 2.4 km s Mpc (Riess et al. 2011) History and Future of the Expansion a/a0 =1/(1 + z)

different pasts and futures depend on gravity and the contents of the Universe

d /c ∼ L History and Future of the Expansion

a/a0 =1/(1 + z)

different pasts and futures depend on gravity and the contents of the Universe

“nearby” supernovae

d /c ∼ L to see these curves diverge, we need to look farther back in time: distant (high-redshift) type Ia supernovae Finding Distant Supernovae ESSENCE Survey

! 5 year project on 4m telescope at CTIO in Chile

! Started in 2002/3. 30 half-nights

! Wide field images in RI bands, 32 fields, every 4 nights, 11 "-deg

! Same-night detection of SNe

! Spectroscopy

! Magellan, Keck, Gemini telescopes

! Goal is ~200 SNe, 0.2

! Distances to 2% in !z=0.1 bins

CTIO 4m Image Subtraction

(High-z Supernova Team)

10 A Distant Type Ia Supernova, z = 1.3 The Mighty Aphrodite

ACS z-band 1 orbit

2002 August 01 The Mighty Aphrodite

ACS z-band 1 orbit

2002 September 22 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 01 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 05 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 10 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 20 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 30 The Mighty Aphrodite

ACS z-band 1 orbit

2002 October 31 The Mighty Aphrodite

ACS z-band 1 orbit

2002 November 17 The Mighty Aphrodite

ACS z-band 1 orbit

2002 November 25 High-Z SN Search Team Supernova Cosmology Project Riess et al. (1998) Perlmutter et al. (1999) High-Z SN Search Team Supernova Cosmology Project Riess et al. (1998) Perlmutter et al. (1999) 2011 Nobel Prize in Physics from the New York Times

OBSERVATIONAL EVIDENCE FROM SUPERNOVAE FOR AN ACCELERATING UNIVERSE AND A COSMOLOGICAL CONSTANT THE ASTRONOMICAL JOURNAL, 116:1009È1038, 1998 September ADAM G.RIESS,1 A(LEXEI1998. TheV.F AmericanILIPPENKO Astronomical,1 PETER Society.CHALLIS All,2 rightsALEJANDRO reserved. PrintedCLOCCHIATTI in U.S.A. ,3 ALAN DIERCKS,4 PETER M.GARNAVICH,2 RON L.GILLILAND,5 CRAIG J.HOGAN,4 SAURABH JHA,2 ROBERT P. KIRSHNER,2 B.LEIBUNDGUT,6 M. M.PHILLIPS,7 DAVID REISS,4 BRIAN P.SCHMIDT,8,9 ROBERT A. SCHOMMER,7 R. CHRIS SMITH,7,10 J.SPYROMILIO,6 CHRISTOPHER STUBBS,4 NICHOLAS B. SUNTZEFF,7 AND JOHN TONRY11 Received 1998 March 13; revised 1998 May 6 OBSERVATIONAL EVIDENCE FROM SUPERNOVAE FOR AN ACCELERATING UNIVERSE AND A COSMOLOGICAL CONSTANT ABSTRACT We present spectral and photometric observations of 10 Type Ia supernovae (SNe Ia) in the redshift range 0.16 ¹ z ¹ 0.62. The luminosityADAM distancesG.RIESS of these,1 A objectsLEXEI areV. determinedFILIPPENKO by methods,1 PETER that employCHALLIS,2 ALEJANDRO CLOCCHIATTI,3 ALAN DIERCKS,4 relations between SN Ia luminosityPETER andM. lightG curveARNAVICH shape., Combined2 RON L. withGILLILAND previous data,5 C fromRAIG ourJ.HOGAN,4 SAURABH JHA,2 ROBERT P. KIRSHNER,2 High-z Supernova Search Team and recent results by Riess et al., this expanded set of 16 high-redshift supernovae and a set of 34 nearby supernovaeB.LEIBUNDGUT are used,6 to M. place M. constraintsPHILLIPS on,7 the D followingAVID REISS cosmo-,4 BRIAN P.SCHMIDT,8,9 ROBERT A. SCHOMMER,7 logical parameters: the Hubble constant(H ), the mass densityR. () C),HRIS the cosmologicalSMITH, constantJ.SPYROMILIO (i.e., the , CHRISTOPHER STUBBS, 0 M 7,10 6 4 vacuum energy density,) ), the deceleration parameter (q ), and the dynamicalNICHOLAS age ofB. theSUNTZEFF universe (t,). AND JOHN TONRY The distances of the high-redshift" SNe Ia are, on average, 10%0 È15% farther than expected in a low mass07 11 Received 1998 March 13; revised 1998 May 6 density()M \ 0.2) universe without a cosmological constant. Di†erent light curve Ðtting methods, SN Ia subsamples, and prior constraints unanimously favor eternally expanding models with positive cosmological constant (i.e.,) [ 0) and a current acceleration of the expansion (i.e.,q \ 0). With no prior " 0 constraint on mass density other than)M º 0, the spectroscopically conÐrmed SNe Ia are statistically consistent withq \ 0 at the 2.8 p and 3.9 p conÐdence levels, and with) [ 0 at the 3.0 p and 4.0 p 0 " conÐdence levels, for two di†erent Ðtting methods, respectively. Fixing a ““ minimal ÏÏ mass density,ABSTRACT)M \ 0.2, results in the weakest detection,) [ 0 at the 3.0 p conÐdence level from one of the two methods. We" present spectral and photometric observations of 10 Type Ia supernovae (SNe Ia) in the redshift For a Ñat universe prior()M ] ) \ 1), the spectroscopically conÐrmed SNe Ia require) [ 0 at 7 p and 9 p formal statistical signiÐrangecance" for 0.16 the¹ twoz di†erent¹ 0.62.Ðtting The methods. luminosity A universe distances closed" by ordinary of these objects are determined by methods that employ matter (i.e.,)M \ 1) is formallyrelations ruled out at between the 7 p to SN 8 p con IaÐdence luminosity level for the and two lightdi†erent curveÐtting shape. Combined with previous data from our methods. We estimate the dynamical age of the universe to be 14.2 ^ 1.7 Gyr including systematic uncertainties in the current CepheidHigh- distancez scale.Supernova We estimate Search the likely Team e†ect and of several recent sources results of system- by Riess et al., this expanded set of 16 high-redshift atic error, including progenitorsupernovae and metallicity and evolution, a set of extinction, 34 nearby sample supernovae selection bias, are local used to place constraints on the following cosmo- perturbations in the expansion rate, gravitational lensing, and sample contamination. Presently, none of these e†ects appear to reconcilelogical the data parameters: with) \ 0 and theq º Hubble0. constant(H ), the mass density ()M), the cosmological constant (i.e., the " 0 0 Key words: cosmology: observationsvacuum È supernovae: energy density, general ) ), the deceleration parameter (q ), and the dynamical age of the universe (t ). The distances of the high-redshift" SNe Ia are, on average, 10%0 È15% farther than expected in a low mass0 density()M \ 0.2) universe without a cosmological constant. Di†erent light curve Ðtting methods, SN Ia subsamples, and prior constraints unanimously favor eternally expanding models with positive cosmo- ÈÈÈÈÈÈÈÈÈÈÈÈÈÈÈ 1. INTRODUCTION 1 Department of Astronomy, University oflogical California at constant Berkeley, (i.e.,) [ 0) and a current acceleration of the expansion (i.e.,q \ 0). With no prior Berkeley, CA 94720-3411. This" paper reports observations of 10 new high-redshift 0 2 Harvard-Smithsonian Center for Astrophysics,constraint 60 Garden Street, on massType density Ia supernovae other (SNe than Ia) and) theº values0, the of the spectroscopically cosmo- conÐrmed SNe Ia are statistically Cambridge, MA 02138. M consistent withq logical\ 0 at parameters the 2.8 derivedp and from 3.9 them.p con TogetherÐdence with levels, the and with) [ 0 at the 3.0 p and 4.0 p 3 Departamento de Astronom• a y Astrof• sica, PontiÐcia Universidad 0 " Cato lica, Casilla 104, Santiago 22, Chile. conÐdence levels, forfour two high-redshift di†erent supernovaeÐtting previously methods, reported respectively. by our Fixing a ““ minimal ÏÏ mass density, ) \ 4 Department of Astronomy, University of Washington, Box 351580, High-z Supernova Search Team(Schmidt et al. 1998; M Seattle, WA 98195. 0.2, results in theGarnavich weakestet detection, al. 1998a) and) two[ others0 at (Riess the et 3.0 al. 1998b),p conÐdence level from one of the two methods. 5 Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, the sample of 16 is now large" enough to yield interesting MD 21218. For a Ñat universe prior()M ] ) \ 1), the spectroscopically conÐrmed SNe Ia require) [ 0 at 7 p 6 European Southern Observatory, Karl-Schwarzschild-Strasseand 9 formal 2, statisticalcosmological signi resultsÐcance of" high for statistical the two signi di†erentÐcance. Con-Ðtting methods. A universe closed" by ordinary D-85748 Garching bei Mu nchen, Germany. p Ðdence in these results depends not on increasing the 7 Cerro Tololo Inter-American Observatory,matter National (i.e., Optical)M \sample1) is formallysize but on improving ruled out our understanding at the 7 p ofto system- 8 p conÐdence level for the two di†erent Ðtting Astronomy Observatories, Casilla 603, La Serena, Chile. NOAO is oper- atic uncertainties. ated by the Association of Universities for Researchmethods. in Astronomy, We Inc., estimate the dynamical age of the universe to be 14.2 ^ 1.7 Gyr including systematic uncer- under cooperative agreement with the National Sciencetainties Foundation. in the currentThe Cepheid time evolution distance of the cosmic scale. scale We factor estimate depends on the likely e†ect of several sources of system- 8 Mount Stromlo and Siding Spring Observatories, Private Bag, the composition of mass-energy in the universe. While the Weston Creek, ACT 2611, Australia. atic error, includinguniverse progenitor is known to contain and a metallicity signiÐcant amount evolution, of ordi- extinction, sample selection bias, local 9 Visiting Astronomer, Cerro Tololo Inter-American Observatory. perturbations in thenary expansion matter,)M, rate, which gravitational decelerates the expansion, lensing, and its sample contamination. Presently, none of 10 Department of Astronomy, University of Michigan, 834 Dennison dynamics may also be signiÐcantly a†ected by more exotic Building, Ann Arbor, MI 48109. these e†ects appear to reconcile the data with0 and q 0. 11 Institute for Astronomy, University of Hawaii, 2680 Woodlawn forms of energy. Preeminent among) these\ is a possible º Drive, Honolulu, HI 96822. energy of the vacuum() ), EinsteinÏs ““ cosmological" con-0 Key words: cosmology: observations" È supernovae: general 1009

ÈÈÈÈÈÈÈÈÈÈÈÈÈÈÈ 1. INTRODUCTION 1 Department of Astronomy, University of California at Berkeley, Berkeley, CA 94720-3411. This paper reports observations of 10 new high-redshift 2 Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Type Ia supernovae (SNe Ia) and the values of the cosmo- Cambridge, MA 02138. 3 Departamento de Astronom• a y Astrof• sica, PontiÐcia Universidad logical parameters derived from them. Together with the Cato lica, Casilla 104, Santiago 22, Chile. four high-redshift supernovae previously reported by our 4 Department of Astronomy, University of Washington, Box 351580, High-z Supernova Search Team(Schmidt et al. 1998; Seattle, WA 98195. Garnavichet al. 1998a) and two others (Riess et al. 1998b), 5 Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218. the sample of 16 is now large enough to yield interesting 6 European Southern Observatory, Karl-Schwarzschild-Strasse 2, cosmological results of high statistical signiÐcance. Con- D-85748 Garching bei Mu nchen, Germany. Ðdence in these results depends not on increasing the 7 Cerro Tololo Inter-American Observatory, National Optical sample size but on improving our understanding of system- Astronomy Observatories, Casilla 603, La Serena, Chile. NOAO is oper- atic uncertainties. ated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation. The time evolution of the cosmic scale factor depends on 8 Mount Stromlo and Siding Spring Observatories, Private Bag, the composition of mass-energy in the universe. While the Weston Creek, ACT 2611, Australia. universe is known to contain a signiÐcant amount of ordi- 9 Visiting Astronomer, Cerro Tololo Inter-American Observatory. nary matter,) , which decelerates the expansion, its 10 Department of Astronomy, University of Michigan, 834 Dennison M Building, Ann Arbor, MI 48109. dynamics may also be signiÐcantly a†ected by more exotic 11 Institute for Astronomy, University of Hawaii, 2680 Woodlawn forms of energy. Preeminent among these is a possible Drive, Honolulu, HI 96822. energy of the vacuum() ), EinsteinÏs ““ cosmological con- " 1009 2011 Nobel Prize in Physics from the New York Times

ÈÈÈÈÈÈÈÈÈÈÈÈÈÈÈ 1. INTRODUCTION 1 Department of Astronomy, University of California at Berkeley, Berkeley, CA 94720-3411. This paper reports observations of 10 new high-redshift 2 Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Type Ia supernovae (SNe Ia) and the values of the cosmo- Cambridge, MA 02138. 3 Departamento de Astronom• a y Astrof• sica, PontiÐcia Universidad logical parameters derived from them. Together with the Cato lica, Casilla 104, Santiago 22, Chile. four high-redshift supernovae previously reported by our 4 Department of Astronomy, University of Washington, Box 351580, High-z Supernova Search Team(Schmidt et al. 1998; Seattle, WA 98195. Garnavichet al. 1998a) and two others (Riess et al. 1998b), 5 Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218. the sample of 16 is now large enough to yield interesting 6 European Southern Observatory, Karl-Schwarzschild-Strasse 2, cosmological results of high statistical signiÐcance. Con- D-85748 Garching bei Mu nchen, Germany. Ðdence in these results depends not on increasing the 7 Cerro Tololo Inter-American Observatory, National Optical sample size but on improving our understanding of system- Astronomy Observatories, Casilla 603, La Serena, Chile. NOAO is oper- atic uncertainties. ated by the Association of Universities for Research in Astronomy, Inc., under cooperative agreement with the National Science Foundation. The time evolution of the cosmic scale factor depends on 8 Mount Stromlo and Siding Spring Observatories, Private Bag, the composition of mass-energy in the universe. While the Weston Creek, ACT 2611, Australia. universe is known to contain a signiÐcant amount of ordi- 9 Visiting Astronomer, Cerro Tololo Inter-American Observatory. nary matter,) , which decelerates the expansion, its 10 Department of Astronomy, University of Michigan, 834 Dennison M Building, Ann Arbor, MI 48109. dynamics may also be signiÐcantly a†ected by more exotic 11 Institute for Astronomy, University of Hawaii, 2680 Woodlawn forms of energy. Preeminent among these is a possible Drive, Honolulu, HI 96822. energy of the vacuum() ), EinsteinÏs ““ cosmological con- " 1009 Date: Tue, 7 Oct 1997 15:37:21 -0700 From: [email protected] (adam riess) To: Saurabh Jha, Peter Garnavich, Brian Schmidt, Alan Diercks, Pete Challis, David Reiss Subject: ...2 shopping days left... a grad student’s view Wow! Can't believe its almost here! The Deadline (emphasis on "Dead") (from the trenches)

That's right happy data reducers, only 2 days left to turn in your completed relative SN photometry. Date: Tue, 7 Oct 1997 15:37:21 -0700 From: [email protected] (adam riess) To: Saurabh Jha, Peter Garnavich, Brian Schmidt, Alan Diercks, Pete Challis, David Reiss Subject: ...2 shopping days left... a grad student’s view Wow! Can't believe its almost here! The Deadline (emphasis on "Dead") (from the trenches)

That's right happy data reducers, only 2 days left to turn in your completed relative SN photometry.

Date: Fri, 10 Oct 1997 10:24:10 -0700 From: [email protected] (adam riess) To: Saurabh Jha, Peter Garnavich, Brian Schmidt, Alan Diercks, Pete Challis, David Reiss Subject: many thanks

Dear Hard Workers,

Thanks to all who sent in their light curves yesterday! I received everyone's except for one person (you know who you are) and when they send their's in, I will have them all (so "you know who you are" get moving).

Thanks Again to All -Adam Date: Tue, 7 Oct 1997 15:37:21 -0700 From: [email protected] (adam riess) To: Saurabh Jha, Peter Garnavich, Brian Schmidt, Alan Diercks, Pete Challis, David Reiss Subject: ...2 shopping days left... a grad student’s view Wow! Can't believe its almost here! The Deadline (emphasis on "Dead") (from the trenches)

That's right happy data reducers, only 2 days left to turn in your completed relative SN photometry.

Date: Fri, 10 Oct 1997 10:24:10 -0700 From: [email protected] (adam riess) To: Saurabh Jha, Peter Garnavich, Brian Schmidt, Alan Diercks, Pete Challis, David Reiss Subject: many thanks

Dear Hard Workers,

Thanks Again to All -Adam moral: to win a Nobel Prize, trick your students into working harder! !"#$!#%&$'($)*+',#-.$/011'#-$203#1$4$+0565#5,(.$#&%'789$

A. Filippenko, Berkeley, CA, 1/10/1998 10:11am: “Adam showed me fantastic plots before he left for his wedding. Our data imply a non-zero cosmological constant! Who knows? This might be the right answer.”! B. Leibundgut, Garching, Germany, 1/11/1998: 4:19am “Concerning a cosmological constant I'd like to ask Adam or! anybody else in the group, if they feel prepared enough to defend the answer. There is no point in writing an article, if we are not very sure we are getting the right answer.” B. Schmidt, Australia, 1/11/1998: 7:13pm “I agree our data imply a cosmological constant,! but how confident are we in this result? I find it very perplexing….”! R. Kirshner Santa Barbara, CA 1/12/1998 10:18am: “I am worried. In your heart you know [the cosmological ! constant] is wrong, though your head tells you that you don’t care and you’re just reporting the ! observations…It would be silly to say ‘we MUST have a nonzero [cosmological constant]’ only to! retract it next year.”

M. Phillips Chile, 1/12/1998, 04:56 am:“…As serious and responsible scientists (ha!), we all know that it is FAR TOO EARLY to be reaching firm conclusions about the value of the cosmological constant”!

J. Tonry, Hawaii, 1/12/1998, 11:40 am:“…who remembers the detection of the magnetic monopole and other gaffs?…on the other hand, we should not be shy about getting our results out …”!

A. Filippenko 1/12/1998, 12:02 pm:“If we are wrong in the end, then so be it. At least we ran in the race.” A. Riess Berkeley, CA 1/12/1998 6:36pm: “The results are very surprising, shocking even. I have avoided telling anyone about them because I wanted to do some cross checks (I have) and I wanted to get further into writing the results up…The data require a nonzero cosmological constant! Approach these results not with your heart or head but with your eyes. We are observers after all!”

A. Clocchiatti, Chile 1/13/1998 07:30pm: “If Einstein made a mistake with the cosmological constant…Why couldn’t we?”!

N. Suntzeff Chile 1/13/1998 1:47pm: “I really encourage you [Adam] to work your butt off on this. We need to be careful…If you are really sure that the [cosmological constant] is not zero—my god, get it out! I mean this seriously—you probably never will have another scientific result that is more exciting come your way in your lifetime.”! History and Future of the Expansion History and Future of the Expansion

different pasts and futures depend on gravity and the contents of the Universe

History and Future of the Expansion

different pasts and futures depend on gravity and the contents of the Universe DENSITY ≠ DESTINY

cosmic acceleration! the High-Z SN Search Team Stockholm, December 10th 2011 “for the discovery of the accelerating expansion of the Universe through observations of distant supernovae” The Accelerating Universe

from http://www.hubblesite.org The Accelerating Universe The Accelerating Universe

so what causes cosmic acceleration? What Causes Cosmic Acceleration? What Causes Cosmic Acceleration?

dark energy! What Causes Cosmic Acceleration?

dark energy!

the cosmological constant? What Causes Cosmic Acceleration?

dark energy!

quintessence? other dimensions?

the cosmological constant? modiﬁcations to general relativity? Things that Concern Theorists

why now? why so small? A Bleak Future? Maybe, Maybe Not

a cosmological constant leads to exponential growth

why now? why so small? but dynamic dark energy means many outcomes are possible, including a Big Crunch (recollapse) or a Big Rip (in ﬁnite time) A Bleak Future? Maybe, Maybe Not

a cosmological constant leads to exponential growth

more precise measurements here are still needed to help answer these questions!

why now? why so small? but dynamic dark energy means many outcomes are possible, including a Big Crunch (recollapse) or a Big Rip (in ﬁnite time) SN Ia Hubble Diagram

Calán/Tololo: Hamuy et al. (1996) 45 CfA1: Riess et al. (1999) CfA2: Jha et al. (2006) SNLS: Astier et al. (2006) ESSENCE: Wood−Vasey et al. (2007) HST: Riess et al. (2007) SDSS: Kessler et al. (2009) 40 CfA3: Hicken et al. (2009)

Hubble Diagram of 399 Type Ia Supernovae 35 SNANA/MLCS2k2 distance modulus [mag] ΩM = 0.3, ΩΛ = 0.7 ΩM = 0.3, ΩΛ = 0.0 ΩM = 1.0, ΩΛ = 0.0 0.4 0.2 0.0 −0.2 −0.4 residual [mag] 0.01 0.02 0.05 0.1 0.2 0.5 1.0 2.0 redshift for the ﬁrst time we have a continuous expansion history measured from SN to z > 1 110 RIESS ET AL. Vol. 659 The Epoch of Deceleration Riess et al. (2007)

Fig. 6.—MLCS2k2 SN Ia Hubble diagram. SNe Ia from ground-based dis- coveriesHST-discovered in the gold sample are shown supernovae as diamonds, HST-discovered at SNevery Ia are high Fig. 7.—Uncorrelated estimates of the expansion history. Following the method of Wang & Tegmark (2005) we derive 3, 4, or 5 independent measure- shown as filled symbols. Overplotted is the best fit for a flat cosmology: M ¼ ments of Hzfrom the gold sample using nÁz 40, 20, and 15, respectively. 0:redshift27, Ã 0:73. Inset show: Residual the Hubble Universe diagram and models decelerated after subtracting ðÞ ¼ empty universe¼ model. The gold sample is binned in equal spans of nÁz 6 The bottom panel shows the derived quantitya ˙ versus redshift. In this plane a pos- wherebeforen is the number the of SNecurrent in a bin and Áaccelerationz is the redshift range of the bin.¼ itive or negative sign of the slope of the data indicates deceleration or acceleration of the expansion, respectively. independent of the cosmological model. Such information may show a model with recent acceleration (q0 0:6) and previ- be more general and of longer lasting value than constraints on ¼À any single, specific model of dark energy. ous deceleration dq/dz 1:2, where qz q0 zdq/dz,which is a good fit to the data.¼ ðÞ¼ þ Following Wang & Tegmark (2005; see also Daly & Djorgovski In Figure 8 we demonstrate the improvement in the measure 2004), we transform the gold sample of luminosity distances to comoving coordinate distances, rz,as of Hz at z > 1 realized from the addition of the new SNe Ia, ðÞ presentedðÞ here, to the sample from R04: we have reduced the un- certainty of Hzat z > 1 from just over 50% to just under 20%.20 1 =5 5 rz 10 0 À : 5 We also repeatedðÞ the analysis of R04 in which the deceleration ðÞ¼2997:91 z ð Þ 2 1 ðÞþ parameter, qz a¨ /a /H z dHÀ z /dt 1, is param- ðÞðÞ À ðÞ¼ ½ðÞ À eterized by qz q0 zdq/dz and determined from the data To facilitate calculations we assume spatial flatness (as motivated ðÞ¼ þ by theoretical considerations, i.e., that most inflation models pre- and equation (5). As in R04, we find that the gold set strongly 5 favors a universe with recent acceleration (q0 < 0) and previous dict K < 10À , or by similar resolutions to the ‘‘flatness prob- lem’’), but the following approach can be generalized to allow for nontrivial spatial curvature. After sorting the SNe Ia by redshift, 20 Monte Carlo simulations of the determination of uncorrelated components we define the quantity of Hz show that the increase in precision proceeds as approximately n 2=3, sig- nificantlyðÞ faster than n1=2,wheren is the number of SNe due to the rate of increase ri 1 ri in unique pairs of SNe. xi þ À ; 6 ¼ zi 1 zi ð Þ þ À where the mean value of xi gives an unbiased estimate of the in- verse of Hzat the redshift, zi. As in Wang & Tegmark (2005), we flux averageðÞ the data first (nÁz 1) to remove possible lensing bias. We then calculate the minimum-variance values of Hz in three, four, or five even-sized bins across the sample, with nÁðÞz chosen to be 40, 20, or 15, respectively, to achieve the desired number of bins. In the top panel of Figure 7 we show sets of three, four, or five samplings of Hzversus redshift from the gold sample. As seen, Hz remains wellðÞ constrained until z 1:3, beyond which the SNðÞ sample is too sparse to usefully determine Hz.Forcom- parison, we show the dynamical model of HzðÞderived from 2 2 3 ðÞ Hz H M 1 z with ‘‘concordance’’ values of ðÞ¼ 0 ð ½þ þ ÃÞ M 0:29 and Ã 0:71. In¼ the bottom panel¼ of Figure 7 we show the kinematic quantity a˙ Hz/1 z versus redshift. In the uncorrelated a˙ versus redshift¼ ðÞ space,ðÞþ it is very easy to evaluate the sign of the change in expansion rate independent of the cosmological model. For com- parison we show three simple kinematic models: purely acceler- 2 ating, decelerating, and coasting, with qz a¨ /a /H z Fig. 8.—Same as top panel of Fig. 7 comparing the improvement to the high- 1 ðÞðÞ À ðÞ¼ dHÀ z /dt 1 0:5, 0.5, and 0.0, respectively. We also est redshift measure of Hz due only to the newest HST data, i.e., since R04. ½ðÞ À ¼ À ðÞ 110 RIESS ET AL. Vol. 659 The Epoch of Deceleration Riess et al. (2007)

680 RIESS ET AL. Vol. 607 Fig. 6.—MLCS2k2 SN Ia Hubble diagram. SNe Ia from ground-based dis- coveriesHST-discovered in the gold sample are shown supernovae as diamonds, HST-discovered at SNevery Ia are high Fig. 7.—Uncorrelated estimates of the expansion history. Following the method of Wang & Tegmark (2005) we derive 3, 4, or 5 independent measure- shown as filled symbols. Overplotted is the best fit for a flat cosmology: M ¼ ments of Hzfrom the gold sample using nÁz 40, 20, and 15, respectively. 0:redshift27, Ã 0:73. Inset show: Residual the Hubble Universe diagram and models decelerated after subtracting ðÞ ¼ empty universe¼ model. The gold sample is binned in equal spans of nÁz 6 The bottom panel shows the derived quantitya ˙ versus redshift. In this plane a pos- wherebeforen is the number the of SNecurrent in a bin and Áaccelerationz is the redshift range of the bin.¼ itive or negative sign of the slope of the data indicates deceleration or acceleration of the expansion, respectively. independent of the cosmological model. Such information may show a model with recent acceleration (q0 0:6) and previ- be more general and of longer lasting value than constraints on ¼À any single, specific model of dark energy. ous deceleration dq/dz 1:2, where qz q0 zdq/dz,which is a good fit to the data.¼ ðÞ¼ þ Following Wang & Tegmark (2005; see also Daly & Djorgovski they also provide the first In Figure 8 we demonstrate the improvement in the measure 2004), we transform the gold sample of luminosity distances to

of Hz at z > 1 realized from the addition of the new SNe Ia,

comoving coordinate distances, rz,as constraints on dynamic darkpresented ðÞ here, to the sample from R04: we have reduced the un- ðÞ certainty of Hzat z > 1 from just over 50% to just under 20%.20 energy,1 still =5 consistent5 with rz 10 0 À : 5 We also repeatedðÞ the analysis of R04 in which the deceleration ðÞ¼2997:91 z ð Þ 2 1 theðÞ þcosmological constantparameter, qz a¨ /a /H z dHÀ z /dt 1, is param- ðÞðÞ À ðÞ¼ ½ðÞ À eterized by qz q0 zdq/dz and determined from the data To facilitate calculations we assume spatial flatness (as motivatedFig. 10.—Joint confidence intervals derived from SN samples for a two-parameter model of the equation of state of dark energy, w(z) w0 w0z. For each panel, ðÞ¼ þ ¼ þ constraints fromandan SN s equationample are combin (5).ed with th Ase prior in, M R04,0:27 we0:04, to findyield the thatindicate thed confid goldence inter setvals. The stronglyposition of a cosmological constant by theoretical considerations, i.e., that most inflation models pre-( 1, 0) is indicated as a filled symbol. The bottom right panel shows¼ the iÆmpact of adding or subtracting a systematic error in distance modulus of 0:05z mag to/from 5 thÀe gold samplefavors. a universe with recent acceleration (q0 < 0) and previous dict K < 10À , or by similar resolutions to the ‘‘flatness prob- lem’’), but the following approach can be generalized to allow forcompelling, empirical case for (or against) a static dark energy current sample of such SNe Ia, although greatly expanded nontrivial spatial curvature. After sorting the SNe Ia by redshift,(i.e., a cosmolo20gical constant). The addition of the silver here, remains small (i.e., <10). Our capacity to constrain si- sample has only a mMonteodest im Carlopact on simulationsthis analysis, ofas theseen determinationin multaneou ofsly uncorrelatedthe full debate componentsd range of cosmological and en- we define the quantity Figure 10 (aoflthHzough ashowcosmo thatlogica thel co increasenstant cros inses precisionto just proceedsvironmenta asl pa approximatelyrameters is therefonre2=l3im,i sig-ted. Consequently, we outside the nnificantlyominðÞal 68% fasterconfid thanence nin1te=2rv,whereal). n is the numberhav ofe c SNehosen dueto t toest thespe ratecific ofan increased narrow questions in the The HST-discovered SNe Ia provide significant leverage in context of well-defined assumptions or in conjunction with ri 1 ri the w -w plinane uniquebecause o pairsf their ofhig SNe.h mean redshift. Figure 10 independent information. It is important to recognize that the x þ ; 6 0 0 i À (top left) shows the w0-w plane without including the HST- conclusions garnered from any analysis cannot furnish a pri- ¼ zi 1 zi ð dÞiscovered objects. The HST-discovered SNe Ia alone increase ori information for a subsequent analysis. Readers should þ À the precision (i.e., reduce the area of the confidence intervals) carefully consider which priors they are using and where by an impressive factor of 1.9, although they account for only they came from before selecting which analysis presented where the mean value of xi gives an unbiased estimate of the in-10% of the sample. Previous studies in support of a dedicated, here provides a relevant incremental gain. space-based mission to measure w0 and w0 have concluded The two most extreme analyses (in the sense of the breadth verse of Hzat the redshift, zi. As in Wang & Tegmark (2005),that an SN Ia sample must extend to z > 1:5 to adequately of their priors) presented here also realize the most significant we flux averageðÞ the data first (nÁz 1) to remove possible lens-break degeneracies in this parameter space (Linder & Huterer gains from the addition of our highest redshift SNe Ia dis- 2003), a conclusion supported by our analysis. The current covered with HST. The kinematic (i.e., cause-independent) in- ing bias. We then calculate the minimum-variance values of Hzrelative dearth of SNe Ia at z > 1 compared to their number at terpretation of SN Ia distances and redshifts (independent of in three, four, or five even-sized bins across the sample, with nÁðÞzz< 1 indicates that significant progress can still be made in all other experiments) is most consistent with two distinct the constraints on w0. Proposals for a Supernova/Acceleration epochs of expansion: a recent accelerated expansion and a chosen to be 40, 20, or 15, respectively, to achieve the desiredProbe (SNAP) or a Joint Dark Energy Mission (JDEM) predict previous decelerated expansion with a transition between the an improved constraint for w0 over our current analysis by a two at z 0:5. This is a generic requirement of a mixed dark number of bins. factor of 3–4, assuming a similar-sized improvement in our matter and dark energy universe, and it may even be a feature In the top panel of Figure 7 we show sets of three, four, or fiveknowledge of M from the Planck satellite and a continued of unrelated cosmological paradigms (which are beyond our ability to reduce systematic errors (Linder & Huterer 2003). scope to consider here). The data are not consistent with many samplings of Hzversus redshift from the gold sample. As seen,However, the current sample is rapidly growing in size, and astrophysical interpretations posited in lieu of dark energy. Hz remains wellðÞ constrained until z 1:3, beyond which thewe may expect progress in our constraints on the nature of Notable examples include the attenuation produced in a uni- ðÞ dark energy in the next few years. verse filled with gray dust at z > 1, or a luminosity evolution SN sample is too sparse to usefully determine Hz.Forcom- that is a simple, monotonic function of redshift. These inter- ðÞ 5. DISCUSSION pretations are robust against the exclusion of any individual parison, we show the dynamical model of Hz derived from SN Ia used in the analyses and therefore represent an im- 2 2 3 ðÞ 5.1. Cosmological Constraints Hz H M 1 z Ã with ‘‘concordance’’ values of provement over the results of Riess et al. (2001). ðÞ¼ 0 ð ½þ þ Þ SNe Ia at z > 1 provide valuable and unique contributions A vacuum-driven metamorphosis model (VCDM) has been M 0:29 and Ã 0:71. to our current understanding of the cosmological model. The proposed by Parker & Raval (1999) to explain the cause of In¼ the bottom panel¼ of Figure 7 we show the kinematic quantity a˙ Hz/1 z versus redshift. In the uncorrelated a˙ versus redshift¼ ðÞ space,ðÞþ it is very easy to evaluate the sign of the change in expansion rate independent of the cosmological model. For com- parison we show three simple kinematic models: purely acceler- 2 ating, decelerating, and coasting, with qz a¨ /a /H z Fig. 8.—Same as top panel of Fig. 7 comparing the improvement to the high- 1 ðÞðÞ À ðÞ¼ dHÀ z /dt 1 0:5, 0.5, and 0.0, respectively. We also est redshift measure of Hz due only to the newest HST data, i.e., since R04. ½ðÞ À ¼ À ðÞ Cosmological results: flat, constant w with additional constraints from CMB, BAO

Nearby + SDSS

Nearby + SDSS + ESSENCE + SNLS +HST w = −0.96 ± 0.06(stat) ± 0.12(syst) 1 Kessler et al. (2009)

€ the future of (understanding) dark energy • cosmology with type Ia supernovae is now systematics limited (~10% precision in w): • progenitors, evolution, standardization, photometric calibration, etc. • constraints on w to a few percent are still feasible with increasing samples, rest-frame infrared observations

• complementary techniques have enormous utility in probing dark energy: • cosmic microwave background • baryon acoustic oscillation + other LSS

• local measurement of H0 (+ with SN Ia) • weak lensing projected dark energy constraints • galaxy clusters + number counts (ﬁgure from lsst.org) STScI-PRC-01-09

A Preposterous, Extravagant, or Just Interesting Universe?

This diagram shows the ingredients that make up the universe. Astronomers now realize that the universe's main ingredient is "dark energy," a mysterious form of energy that exists between galaxies. The next largest constituent is dark matter, which is an unknown form of matter. The rest of the universe consists of ordinary matter. Most of it is locked up in stars and clouds of gas. A tiny fraction of this matter is composed of heavier elements, the stuff of which humans and planets are made. STScI-PRC-01-09

A Preposterous, Extravagant, or Just Interesting Universe?

that’s us! (ugly bags of mostly water) This diagram shows the ingredients that make up the universe. Astronomers now reooh,alize th scary!at the universe's main ingredient is "dark energy," a mysterious form of energy that exists between galaxies. The nwe’reext larg enotst con surestituen t is dark matter, whwhereich is an uthisnkno wisn form of matter. The rest of the universe consists of ordinary matter. Most of it is locked up in stars and clouds of gas. A tiny fraction of this matter is cowempo sdon’ted of h eknowavier ele ments, the stuff of whicwhath hum athisns an dis planets are made.

and we know even less about this!