Image Earle th f yso Universe from the BOOMERanG experiment

Bernardise Pd . 1, P.A.R.Ade2, JJ.Bock3, J.R.Bond4, J.Borrill5'6, A.Boscaleri7, K.Coble8, B.P.Crill9, G.De Gasperis10, G.De Troia1, P.C.Farese8, P.G.Ferreira11, K.Ganga9'12, M.Giacometti1, E.Hivon9, V.V.Hristov9, AJacoangeli1, A.HJaffe6, A.E.Lange9, L.Martinis13, S.Masi1, P.Mason9, P.D.Mauskopf14'15, A.Melchiorri1, L.Miglio16, T.Montroy8, C.B.Netterfield16, E.Pascale7, F.Piacentini1, G.Polenta1, D.Pogosyan4, S.Prunet4, S.Rao17, G.Romeo17, J.E.Ruhl8, F.Scaramuzzi13, D.Sforna1, N.Vittorio10

Dipartimento1 Fisica,di Universitd Romadi Sapienza,La 001852, P.leRoma,Mow A. Italy, 2 Department of Physics, Queen Mary and Westfield College, Mile End Road, London, El 4NS, UK, 3 Jet Propulsion Laboratory, Pasadena, CA, USA, 4 CITA University of Toronto, Canada, NERSC-LBNL,5 Berkeley, USA,CA, 6 Center for Particle Astrophysics, University of California at Berkeley, 301 Le Conte Hall, Berkeley CA 94720, USA, 1IROE CNR,- PanciatichiVia 64,50127 Firenze, Italy, 8 Department of Physics, University of California at Santa Barbara, Santa Barbara, CA 93106, USA, 9 California Institute of Technology, Mail Code: 59-33, Pasadena, CA 91125, USA, 10 Dipartimento di Fisica, Universitd di Roma Tor Vergata, Via della Ricerca Scientifica 1, 00133 Roma, Italy, 1 ] Astrophysics, University of Oxford, Keble Road, 3RH,OX1 UK, 12 PCC, College de France, 11 pi. Marcelin Berthelot, 75231 Paris Cedex 05, France, 13 ENEA Centra Ricerche Frascati,i d . FermiE a , 00044Vi 45 Frascati, Italy, 4 Physics1 Astronomyd an Dept, Cardiff University,, UK 5 Dept1 of Physics Astronomy,d an UMass. Amherst, , USA,MA 16 Departments of Physics and Astronomy, University of Toronto, Canada, 17 Istituto Nazionale di Geofisica, Via di Vigna Murata 605, 00143, Roma, Italy,.

THE CMB AND THE CURVATURE OF THE UNIVERSE

The CMB is the fundamental tool to study the properties of the early universe and of the universe at large scales. In the framework of the Hot Big Bang model, when we look to plasme th f looo e a w d redshif keraa t B en bac a ,e timn th kCM i t o e et ~ th 1000, when the univers ^s 5000ewa 0 times younger, ~ 1000 times time9 hotte ~10 d san rdense r than today. The image of the CMB can be used to study the physical processes there, to infer what happened before alsd studo an ,t backgroune yth d geometr Universer ou f yo . traveB spacn photone CM i billiol 5 r ~e 1 Th efo th f nso years before reachinr gou

CP586, Relativistic Astrophysics: 20th Texas Symposium, edited by J. C. Wheeler and H. Mattel © 2001 American Institut Physicf eo s 0-73 54-0026-1/017$ 18.00 157

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp microwave telescopes. Originally visible and near infrared light, they are converted in a faint glow of microwaves by the expansion of the Universe. Since CMB photons travel so long in the Universe, their trajectories are affected significantly by any large-scale curvature of space. The image of the CMB can thus be used to study the large scale geometry of space. The scale of the acoustic horizon at recombination is the "standard ruler" neede thesr dfo e studies: density fluctuations larger tha horizoe nth frozene nar , while fluctuations smaller than the horizon can oscillate, arriving at recombination in a compressed or rarefied state, and thus producing a characteristic pattern of hot and cold spots in the CMB (see e.g. [1] [2] [3] [4] [5]). In fact temperature th , relatee densite ar e th fluctuation B do t yCM fluctuatione th f so s at recombination through three physical processes: the photon density fluctuations 8Y accompanying the fluctuation of density; the gravitational redshift/blueshift of photons coming from overdense/underdense regions with gravitational potential <|); the Doppler shift produced by scatter of photons by electrons moving with velocity v with the perturbation. In formulas (see e.g.[6]):

where n is the line of sight vector and the subscript r labels quantities at recombination. In the CMB temperature distribution we see a snapshot of the status of the density perturbation t recombinationa s characteristie th : horizoe c e densitth siz th f eo n ni y fluctuations is thus directly translated in a characteristic size in the spots of the CMB. Recombinatio f hydrogeno n happens whe temperature nth e drops below ~ 300(MT, i.e. about 300000 year Bangg scausae Bi afte e Th . th rl horizo abous ni t 300000 light years there. Since the distance travelled by CMB photons is ^ 15 billion light years, and lenghts in the Universe increase by a factor 1000 meanwhile, the angle subtended now by the horizons at recombination is expected to be close to one degree, in a Euclidean Universe typicae Th . l observe dcold angulaan d t spotho re s sizth strongl f eo y depends on the average mass-energy density parameter Q. The presence of mass and energy acts as a magnifying (Q > 1) or demagnifying (Q < 1) lens, producing horizon-sized spots larger or smaller than ~ \° in the two cases, respectively. The image of the CMB is described in statistical terms: the temperature field is expane multipolen di s

The aim are random variables with zero average and ensemble variance < atma*£,m, >= $U'&mm'- represenS Q' e angulae Th th t r power spectru CMBe compute th w f mf o I . e thc e power spectru imagCMBe e peaa th th expec e m f f e multipolet eko o ,w a se o tt \ s^i 200 corresponding to these degree-size spots. The location of the peak will be mainly drive, thuQ sy nb allowin gmeasuremena thif o t s elusive cosmological parametere Th . dependanc t simplpresentle no th \ i fros i n ef i emo Q y favoured cosmological model with significant vacuum energy OA [7] [8], Full spectral data must be compared to spectral models compute f cosmologicao d t frose ma l parameters degeneracied an , s mus takee tb n into account [9], [10] remaint .I s confirmed, however, tha maie tth n driver

158

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp locatioe fo th firsre th t f nacoustio c . valuThie peathun O th retrievef e sca s kesi o b d with good accuracy from the power specturm of an image of the CMB with sub-degree resolution. After recombination, the same density fluctuations driving the acoustic oscillations grow, and form the jerarchy of structures we see in the Universe today. Thus, the image of CMB anisotropies represents also a fundamental tool in the study of the formation of structure Universee th n si . The detailed shape of the CMB power spectrum Q has been computed in a num- ber of scenarios with very high detail. A wide literature is available on the subject as wel publicalls a l y available software ([11], [12] accuratelo )t y comput powee eth r spec- trum cosmologicaf o give t se na l parameter framewore th n si adiabatif ko c inflationary models. More work remains to be done for the general case (for example including isocurvature modes, see e.g. [13]). maie Th n feature, i.e harmonia . c serie f peakso s followin describefirse e gth on t d above, can be understood as follows. The acoustic horizon increases with time, and at some point becomes larger tha givena n perturbation size t thiA . s perture pointth l al ,- bations present in the Universe with that size will start to oscillate. This can be seen as a cosmic synchronization process, which initiates the oscillation of small perturbations befor oscillatioe eth largf no e ones perturbationphase e :th th f eo recombinatioe th t sa n depends on their intrinsic size. Perturbations with size close to the horizon at recom- bination start lasthavd an ,e just enough tim maximuarrive o et th o et m compression, producing the degree-size spots in the CMB, i.e. the first "acoustic" peak in the angular powe anisotropyB r spectruCM e th f m. o Perturbation s with smaller intrinsic size have entere acoustie dth c horizon before arrivd an , recombinatiot ea n afte fula r l compression and a return to the average density. They will produce a CMB temperature fluctuation smaller tha previoue nth s ones, because amon three gth e physical processes producing temperature fluctuations from density ones, onl Dopplee yth r effec effectives i t . This angulae th correspond n i rp powe di t multipole a a B o srt spectruCM e s th large f mo r tha firse nth t acoustic peak. Even smaller perturbations have enough tim compresso et , return to the average density, and then arrive to the maximum rarefaction at recombina- tion, producing a second peak in power spectrum, at multipoles about twice of those of firse th t one. Repeating this reasoning expece ,w harmonita c serie acoustif so c peaksp ,u vero t y small sizes, smaller tha thicknese nth lase th t f scatterinso g surface. Photons dif- fusion, and the fact that many small-size positive and negative perturbations are aligned on the same line of sight, damps the temperature fluctuations measurable in the CMB at very high multipoles. poweshape e th Th f eo r spectru CMBe th , f dependmo Q , severan so l cosmological parameter addition si . IncreasinQ o nt physicae gth l densit f baryonyo s Q&/*2 favours compressions against rarefaction acoustie th n i s c oscillations. Compression peake sar thus enhanced with respect to rarefaction ones. The relative amplitude of the second peak (rarefaction) with respec amplitude th o firse t th t f epeao k (compression thus )i s gooa d measuremen f O^/?o tpowe e 2Th . r spectru f primordiamo l density fluctuations P(k) control generae sth l poweshape th f eo r spectru CMBe inflationare th th f mn o I . y scenari value o als n witP(k) Ak Th f oe . o ~= h n 1 drive amplitude sth highee th f eo r

order peakn s relativ amplitude th o efirst e th t f eoneo onlf .I seconfirs e d yth an t d peae kar observed, increasin abous ha sam e gn t th e effec decreasins a t g Q^/z measuremen2A . f to

159

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp the third peak removes this degeneracy.

CMB AND COSMIC INFLATION

In the previous section we have assumed the presence of density perturbations in the primeval plasma instrumenR COBe 199DM n I . th e 2n th Eo t satellit shows eha n that these perturbations do exist at least at large scales [14]. The angular resolution of DMR was 7°, corresponding to a sensitivity to multipoles between 1 and ~ 20 in the angular power spectru CMBe th powee mf o .Th r spectru characterisa s m ha measure R -DM y db tic power spectrum[15] Q ~ \/(,/(£+ 1). This is consistent with a Harrison-Zeldovich power spectru densitf mo y fluctuation wit n s(1. = P(k) Ak h 20.3)n ± = . Such a spectrum is expected in the simplest inflationary scenarios (see e.g. [16] [17] [18] [19] [20]), where microscopic quantum fluctuations in the very early universe

(~ 10^ s after the big bang) are inflated to cosmological scales by the exponential 6

growt spacf ho 3 e durin earln ga y phase transitio grand-unificatioe th t na n densiterae .Th y fluctuations generated in this way are gaussian and adiabatic (see e.g. [21]). Inflation explains the homogeneity of the CMB at large scales: regions causally disconnected recombinatioe ath t n epoc beed hclosha n n i e causal contac vere th yn i tearl y Universe, befor superluminae eth l inflatio spacef no . Inflation naturally produce flasa t geometrf yo space, stretching any inital curvature, due to the huge expansion factor. The power law spectrum of primordial fluctuations is reminiscent of the initial quantum fluctuations: constraining n through the measurement of the power spectrum is thus a way to test the inflationary hypothesis. However, some inflation variants feature values of n ^ 0.9 or even less (see e.g. [22]) adiabatie Th . c inflationary scenari shows oi cartoon i n formn i fig.l. alternative Th e scenari generatioe th r ofo f densitno y fluctuation bases i s topon do - logical defects (see e.g. [23]) thin .I s scenari gaussian ono n isocurvature fluctuatione sar favored. The peak at i\ is either not present of shifted to higher Is. The analysis of the powe(anB particulan di CM r imag e e spectrugaussianits th th it f f f eo o ro d man y prop- erties) can thus distinguish between the two alternative scenarios for the generation of density fluctuations. problee Th f recoverinmf cosmologicao o t se e gth l parameters fro measuree mth d power spectru bees ha n m Q widel y studie satellite vien dth i f wo e d missionan P sMA Planck, which promise to measure the power spectrum with ~ 1 % accuracy on a very wide range of multipoles, allowing an accurate determination of the main cosmological parameters (see e.g. [25]).

MEASURINB CM IMAGE E GTH TH F EO

The contrast of the image of the CMB is very low (~ 10 ppm). Moreover, atmo- spheri Galactid can c signalmuce b n h sca large anisotropyB r CM tha e nth , depending on wavelength, observe regioy dsk locatiod observere nan th f no e th . p Tryinma o gt CMB anisotroy from the total brightness coming from the sky in a ground-based exper-

160

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp FIGUR . E1 Adiabati c inflationary scenari generatioe th anisotrop B r ofo CM f no y from quantum fluc- tuation vere th yn i searl y universe. Inflation boost microscopie th s c fluctuations producing adiabatic gaussian density fluctuation l cosmologicaal t a s l scales. Fluctuatios larger tha causae nth l horizoe nar effectively frozen, and produce the large scale anisotropy in the CMB measured by the COBE-DMR. Fluctuations enterin horizoe gth n before recombinatio processee nar cosmologicae th y db l plasmd aan oscillat souns ea d waves, thus producin gcharacteristia c power spectrut a B f anisotropmo CM e th n yi sub-horizon scales angulae Th . r scale under which thes seee ear n today depend curvature th n so f eo space.

iment is like trying to map distant galaxies in the visual band during daytime. Detector and istrumental noise is an additional problem in these measurements. l thesal r e Fo reasons year7 2 , s were neede produco dt firse eth t detectioB CM f no anisotropy after its the discovery. In 1992 the DMR instrument on board of the COBE satellite detecte firse th tr t timlargdfo a eB e anisotropscaleCM e sth [14]f e yo Th . resolution of the instrument (~ 1° FWHM) was not sufficient to resolve the degree- scale spots useful to measure the curvature of the Universe. Following this detection, many experiments were carrie witt dou h degre sub-degred ean e resolution firse t.Th clear angulae detectionth n i r pea0 a powei ~ f 20 t s ko a anisotrop B r spectruCM e th y f mo detectorsw arrivene f o e d. witus HEM e hth T based microwave amplifiers [26] [27] were usetelescopT d MA wit e heth operatin t 5000ga f altitudmo Chiln ei e [28] [29]. Spider web bolometers[30] were used at higher frequencies in the test flight of the balloon borne experiment BOOMERanG [31], which was flown in a short flight in Texas in 1997, while TOC takins Owa g data. Both experiments produced convincing evidence peae fo mos~I e rth t k th a 200 t t excitin.Bu g breakthroug recene th s t hmeasuremenwa t of wide, resolved image CMBe th f so , obtaine BOOMERanG-LDB[32e th y db y b d ]an MAXIMA-e th 1 [33] experiments followine th n .I wile gw lresult e focuth n so s froe mth BOOMERanG experiment.

161

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp BOOMERANG

BOOMERanG (Balloon Observations Of Millimeter Extragalactic Radiation and Geo- physics 1.3a s )i m off-axis scanning telescope using fast, ultra-sensitive bolometri- cde tectors in four frequency bands. The telescope scans the sky at nearly constant speed, of the orde f 1°/s,o r alon gwid° L60 scan e = scan t constansa t elevation (either 40°, 45°, or 50°). The main beam is similar to a gaussian with Gbeam ~ 5' at 150 GHz. Different multipoles of the CMB anisotropy are converted by the scan into different audio frequen- cies in the detectors [34], thus avoiding the effects of 1/f noise and other low-frequency disturbances [35]. This allows the detection of the angular power spectrum over a wide range of multipoles (imax ~ ^/®beam \ ^min ~ 1 /LScan) in a single experiment. Two dif- ferent scan speeds (1 dps and 2 dps) have been used during the measurements to detect possible systematic effects due to the transfer function and noise spectrum of the instru- ment. The instrument operates in the stratosphere, avoding the large-scale signals and noise produced by atmospheric emission. The full payload is rotated around the vertical axis to avoid scan synchronous instrumental signals. The azimuth of the center of the scan tracks the azimuth of the selected region to be mapped, and sky rotation produces nicela y crosslinked patter f scansno , very usefu reconstruncfroo p t le mma th y sk e th t time-ordered datamesuremene Th . repeates i t d several time differenn si t days, while instrumene th t drift hundredy sb kilometerf so stratospheris it n si c circumnavigatiof no Antarctica. This allow repetitioe sth measuremente th f no s under very different experi- mental conditions, which is the best way to check for systematic effects contaminating datae r exampleth Fo . groune , everth y dyda environment (ice, sea, rocky areas difs )i - groune th s i dferento s spillove d sidelobee an , th telescopee n ri th f so sensitivA . e null obtainee b tesn ca tcomparin y db mape gth s obtaine differenn di t mapdayse th e f sar I . samee th , ground spillover contaminatio excludede b n nca . BOOMERan G simultaneously GHz0 mapsk 41 e .d th sCom an , 1500 90 - 24 t , ya parison of the maps measured at different frequencies is a powerful tool to test for foregrounds contamination. The two low frequency bands are mainly sensitive to CMB anisotropies, while the two higher frequency bands can be used to monitor atmosperic and interstellar contaminations. 16 detectors have been distributed in the focal plane as shown in fig.2. The location of different detectors in the focal plane has been optimized orden i havo rt e robust confirmatio different a structuref y no sk e t th tim n si e scales. The technical details of the instrument are reported in [36] and in [37]. The main characteristic followss a e sar : • Telescope: off-axis gregorian with cryogenic secondar tertiard yan y • Primary mirror: off-axis, aluminum, 45^ off-axis, 1.3m diameter, f/1.

• 90 GHz detectors: 2, 18' FWHM, be$tNETCMB

• 150 GHz detectors: 6, 10' FWHM, bestNETCMB detectorsz ' FWHMGH 14 0 , :4 24 • , be$tNETCMB

• 410 GHz detectors: 4, 12' FWHM, bestNETCMB • Attitude control: azimuth flywheels, passive pendulation damper Azimut• h 2°/s o scant 1 t sa • Attitude reconstruction: differential GPS, digital sundial sun sensors, laser gyroscopes

162

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp FIGURE 2. The cryogenic focal plane of the BOOMERanG experiment. In the bottom right panel the entrance of the 8 photometers are shown and labeled with the corresponding frequency (in GHz). instrumene Th flows NASA-NSBtwa y nb F from Dec.29, 199 Jan.8,19998o t m K 9 3 t ,a of altitude, circumnavigating Antarctica at a latitude ~ —78°S. 57 million 16 bit samples were acquired for each detector during the flight.

THE MAPS OF THE MICROWAVE SKY

Maximum likelihood sky maps were constructed from the time-streams using the MAD- CAP package [38] and a recursive estimator of instrumental noise [39]. The maps cover ~ 1800 square degrees in one of the best (lowest foreground) sky regions in the south- ern hemisphere. The structures at large angular scales ( •> 10°) have been filtered out to remov noisf effec e instrumenta1/ d e th f ean o t l drifts. These maps, smoothe resoa o dt - lutio' FWHM22 f no , have been publishe [32]dn i fig.n .I sho e (AT243w m wsu ATis0+ o ) and difference ( A7240 — ATiso ) maps obtained from the 150 and 240 GHz chan- nels, expresse temperaturB CM n di e fluctuation units. Degree-size structures uniformly coverin surveyee gth dominane d th are e amapar structurem e t su featur.Th e th f eso dis- appea difference th n ri e map. Thi alreads si cosmologica e prooya th f o f l nature th f eo structures. verIs i t y importan comparo t 0 structuree 24 eth d an 0 s15 visibl , mape 90 th t n sa ei GHz [32]. The structures are very similar in shape, and the relative amplitude of the fluc- tuations is perfectly consistent with the derivative of a 2.73K blackbody, as expected for CMB anisotropy. A simple scatter plot of these signals expressed in CMB temperature fluctuations unit bess slopet sha tfi s very consisten wit confirmin, h1 visuae gth l impres-

163

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp BOOMERANG Preliminary

(Caltech, ENEA, ING, mOE.La Sapienza, QMWC, U€SBUMass,U Toronto) t

FIGURE 3. The sum (top panel) and difference (bottom panel) maps obtained from the 150 and 240 GHz channels of BOOMERanG. CMB temperature units are used for both the channels. For this reason, CMB fluctuations are enhanced in the sum map, while are removed in the difference map. Only non-CMB structures remai difference th n i e map three Th .e circles surround three AGNs presen mape th .n ti

sion above[40]. Masi et al. [41] find that the contamination from the main local fore- ground (i.e. thermal emission from diffuse interstellar dust mea e less )i th f sn o tha % n1 square fluctuation detected at 150 GHz. Moreover, the brightness fluctuations generated by unresolved point-like radio sources can be estimated from existing catalogues ([42]) and is found to be negligible (~ 160(£/1000)2/^2). From these results we conclude that dominane th anisotropyt B featur mape CM th s n si ei , copyin pattere acoustige th th f no c horizons at the last scattering surface. The amplitude of the fluctuations (ATrms ~ SQw^T) is consistent with the level of CMB anisotropy computed in the inflationary adiabatic

164

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp scenario, normalize COBE-DMe th o dt R detection. e MAXIMTh A experimen t s alsa ha toy f ~sk produceo 0.25e p th ma f % o a d GHz0 15 observiny ,b ghiga h latitud regioy Northere eth sk n ni n Hemisphere [33]. BOOMERan MAXIMd Gan A nicely complement each other facn I . t BOOMERanG covers a large region of the sky, producing data with smaller cosmic variance at t ;$ 300, precisiobue th t pointine th n ni g reconstructio presentls ni y limite arcminutes3 o dt . For this reason the maps have been smoothed to 22.5' FWHM and the power spec- tru limitems i t £ o d600t . MAXIMA instead cover smallesa regiony sk r , producing larger error multipolet sa alreads 300ha $ ; st y,bu achieved sub-arcmin precisioe th n i pointing reconstrunction limites i p resolutioe th dma intrinsie d e onlth th an ,y yf b n o c resolution on the telescope at ~ 10' FWHM. The power spectrum correspondingly ex- tends up to I ~ 800. The map from MAXIMA is statistically very similar to the map of BOOMERanG thid san , agreemen vera s ti y important independent confirmatioe th f no results for both experiments.

THE POWER SPECTRUM OF THE CMB

The angular power spectrum[32] has been computed from the maps of BOOMERanG differeno itw n t ways: usin gsimpla e spherical harmonics transform [43 usind an ] g MADCAe th P [38] maximum likelihood algorithm methodo tw e sTh . produce very consistent results fig.n I repor.e 4w angulae th t r power spectru centee th f mo r region of the BOOMERanG map (~ 1% of the sky), together with the data from COBE-DMR and from the recently published MAXIMA, CBI [44] and BIMA [45] experiments. Ther severae ear l features immediately evident from these data levee shapd Th . an l e of fluctuation consistens si t wit hconstana t large leveth t ela scales sample COBEy db - DMR, plu acoustio t peas a e kdu c horizon effect t scalea s s aroun degreee d on sec e .Th - ond peak (and the higher order ones) has not been detected yet, but there is significant power detecte multipolet da s betwee600d an . Ther0 ndampe30 a s ei d tai multipolet la s £ 1000, as expected by photon diffusion and line of sight averaging effects at recom- bination. This behaviou remarkabln i s ri e agreement wit simple hth e theory presented in section 2 and we can start to say that the big picture is correct. A word of caution neededs i , however presence Th . secona f eo d pea consistens ki t wit currene hth t data, but it has not been observed yet. Several experiment promise a detection soon, including combinee th d analysi channel2 1 f o s BOOMERann i s G with refined pointing recon- struction, the second flight of MAXIMA or from the interferometers (DASI [46], CBI [47], VSA [48] etc.) currently getting data. Such detection will be the final proof that Universe th e underwen phast ho a te with acoustic oscillations tha d structure ,an tth e ew Universe th n i e e se toda grows ywa t plasmn ho afte e ath r phase through gravitational instability from the same primordial fluctuations. Several experimental problems are immediately evident from the power spectrum shown in fig.4. The data at 50 ;$ t ,$, 300 are limited in precision by cosmic variance. This will improve with larger surveysArcheope th n i s ,a s [49 TOPHAd ]an T [50] balloon experiments and in the full sky survey of the MAP experiment. Significant calibration errors are still present in the same data sets. A reliable standard of calibration is still

165

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 3U J 80 '. o COBE * BOOMERanG 7n 0 MAXIMA-" 1 J% _ s^ 60 . * CBI IT l^ zL BIMA d_^ 50 ; T _ CM ^40 " I T 11 ^T^ i -3 0JlTjjll - : ^ 20 T I I ^ -

10 -1 T-

n J ...... 1 ...... 1 ...... 1 . .... 10 100 1000 multipole I

FIGURE 4. Recently published measurements of the angular power spectrum of the cosmic microwave background

non-trivia thest a l e wavelengths Dipole th d ean , calibration use BOOMERann di n Gca affectee b scay db n synchronous effects difficul monitoo t t betteo rt r [51]thae % n10 .Th data at 300 <, i <, 600 are still detector noise limited. In the case of BOOMERanG, this will improve with the simultaneous analysis of all the 12 detectors sensitive to CMB anisotropy (the published power spectrum has been computed from the best detector alone). BOOMERane Th G results have been obtained fro mpreliminara y pointing solution. Jitter in the telescope pointing is a relevant concern, and we are currently working on the developmen refinea f to d pointing solution maie .Th n effec pointinf o t g jitte spreao t s ri d the equivalent beam of the telescope. In fig.5 we compute the effect of changing the equivalen tmos e beath n tm o inportan t features detecte spectrume th n di locatioe :th f no the peak, its amplitude, and the ratio between the general level of fluctuations at I > 300 and the amplitude of the peak. It is evident from the figure that the measurement of the locatio peae th kf n o (controllin measuremene gth vers i ) y Q robust f o t , whil othee eth r two quantities (controlling for example the measurement of Q^/z) are less robust.

The present data from the interferometers at I ^ 600 suffer for2 lack of spectral resolu- tion and are still single frequency. Moreover, only differential maps have been produced, presence duth o et f significaneo t ground spillove absolute th n ri e maps. Thi goins si g to improve a lot with the analysis of the larger dataset already acquired and with future system developments. significane Th t mismatch betwee lowese nth t multipoles sample interferometriy db c measurements and the highest multipoles measured by BOOMERanG and MAXIMA is reduce leso dt s tha onca beane 2 eth mpartiae th erro d l ran overla £-bande th f pe o sar taken into account. It is, however, a possible indication of a calibration systematic still

166

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp I——I——I——I——I——I

--0.26 - 10 12 14 beam FWHM (arcmin)

FIGUR . E 5 beae Effec th mf o tFWH mose th Mtn o importan t characteristic powee th f so r spectrum measured by BOOMERanG. The nominal FWHM is 10 arcmin.

unaccounted for. finae Th l measurements will arriv months(launcP w fe e Planca witMA d n he i )an h th k (launch in 2007) satellite experiments. In the following we focus on what we can already say abou curvaturee tth .

COSMOLOGY FRO ANISOTROPB MCM Y

positione Th , amplitud peae widtd th e an kf ho eviden fig. n i consistente 4ar t wite hth general adiabatic inflationary scenario, while the simplest models based on topological date wellth s at a fi defectt . no o sd Using a naive quadratic fit estimator for the MAXIMA and BOOMERanG data we find thapeae th tlocates ki t multipolda BOOMERanr fo (19) = 6 ei 7± G (10, bandcenters from 50 to 300, data from [32]) , I = (233 ± 33) for MAXIMA (10, bandcenters from 55 to 300, data from [33]), t = (195 ± 8) for the combination of the two datasets (10, bandcenters from 65 to 296, data from [52]). These results can shift a little bit higher if one assumes skewed power spectra similar to the adiabatic scenario ones [53]. This locatio maximue th f no mconsistens i t wit fiaha t geometrt no s spacei f yo t i t ,bu univocally relate densite th do t y vanishin n paramete no allo e a w r wf i fo g rO cosmologi - cal constant [8]. Moreover locatioe , th maximue th f n o con o t bese my -th alontwa t no es i strain Q, and there is much more information contained in the datasets. Jaffe et al.[52] carried out a full Bayesian analysis on the combined BOOMERanG, MAXIMA and 2 2 COBE-DMR datasets, constraining simultaneousl parametere yth s Q, O^/z j s Qn , ch .

167

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp It mus stressee b t d that this kin f analysido s assume adiabatin sa c inflationary model. Moreover, it is very important to specify in detail the prior distributions assumed for parameterse eacth f ho . Thi especialls i y importan poweB case CM r th f spectrueo n ti m measurements, sinc importann ea t geometrical degenerac presens yi t [9] differeno s , t combination parametere th f so s produce very similar powe rconfi % spectra95 -e Th . dence interval , takinQ r sfo g into accoun degeneraciese th l tal , range from (0.88 — 1.12) to (0.97— 1.35) depending on the assumed priors and parametrizations [52, 10, 24]. This strongly suggests a flat geometry of the Universe, and at least implies, with 95% confidence curvatura , e length

larger than 8.3 to 2.9 times the Bubble length. Needles sayo st , having demonstrate dvers thaha y ~ 1 O timportan t cosmological consequences. According to several independent measurements and methods, the matter density parameter O^ is significantly smaller than unity and close to 30% [54, 55, 56, 57, 58]. This means that about 70% of the mass and energy present in the Universe must be in a different form: not ordinary matter, not dark matter. The presence of dark energy (a form of repulsive energy with negative pressure) interestinn a s i g hypothesi solvo st puzzlee eth presencs It . bees eha n independently proposed [7,59,60,61,62] to drive the acceleration of the expansion rate of the Universe hypotised to explain the observation of distant supernovae [63, 64]. This hypothesis is still widely debated, the main interpretation problem remaining the lack of a convincing particle physics darkmodee th r , negativlfo e pressure for energf mo y required. same Inth e adiabatic perturbations framework, BOOMERan MAXIMd Gan A con- strain the slope of the initial power spectrum of density perturbations ns in the range 1.01 ±0.17 (95% confidence) [52]. In addition to the O measurement, tese results for ns are also consistent and support the simplest inflationary scenarios [65, 19, 66]. The third parameter constrained by BOOMERanG and MAXIMA is the density of baryons powee th n .I r spectrum this parameter control relative sth e amplitud firse th tf eo secone peath o kt d one. Fro powee mth r spectru mf fig.4o , this rati constraines oi o dt be largea ~2 r than expecte standara r dfo d primordial nucleosynthesis (BBN) (O^/z= 2 2 (0.019 ±0.002) from [67], Qbh = (0.020 ± 0.002) from [68]). This suggests a high physical density of baryons. Depending on the assumed priors, 95% intervals for Q^/z2 ranging from (0.019-0.045) to (0.026-0.048) are obtained from the CMB power spectrum data. These results are only marginally overlapping the BBN one, but given the orthogonalit systematie methode th th f f o yo d san c errors woulI , d rather spea goof ko d overall consistency of the model. It must be stressed, in fact, that the density of baryons enters in the two observable in completely different ways. It controls the primordial abundance of light elements due to its effect in the nuclear reactions happening in e firsth t minute bangg sbi afte t e alsI . th ro affect e densitth s y oscillations (acoustic waves) producin anisotropieB gCM s about 300000 year bangg s bi afte e . th Completelr y different physical phenomen t completela y different regime earle th yn si historf yo the Universe require the density of baryons to be the same to within a factor 1.5: this should be considered a wonderfull success of cosmology. New physics will be required

168

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp only if future, more precise measurements will confirm and increase the present ~ 2a disagreement. Also, note that frequentist methods poin statisticaa o t l consistence th f yo two results [69].

BEYON POWEE DTH R SPECTRUM

At this point the next step is obvious: we should check if the detected temperature fluctuations are gaussian distributed, another prediction of inflation. This is very difficult testo t , since instrumental effect masn ca s k small cosmi gaussianitiesn cno n ca d an , produce subtle non gaussianities as well. Moreover, different methods probe different gaussianitn kindno f so y (see e.g. [70 referenced ]an s therein) realitA . y check, simply using the 1 point distribution of the BOOMERanG map, has been published in [40]: detectee th d pixel temperature fluctuations (normalizen i square m th su o d t e e rooth f to quadratur variancy sk f instrumened o ean t noise varianc than ei t pixel indeee ar ) d very precisely gaussian distributed. This is only a starting point in the demonstration of the gaussian character of the CMB, since the central limit theorem helps a lot in this kind of test. However, it is at least reassuring the fact that we do not detect deviations from gaussianit date th t hig aa n yi h latitude GHz0 15 t ,sa while deviation evidene se ar th n i t data at lower galactic latitudes at 150 GHz, and at all the latitudes in the dust dominated map at 410 GHz. A detailed non gaussianity analysis is underway. The MAXIMA team has already published a first analysis of gaussianity of the CMB map [71]. The main difficult coursef o , yis separato ,t e with high confidence small instrumental effects and foreground contaminations from real non-gaussian signatures in the CMB, if any. Realistic montecarlo simulation asseso t onl e y statisticae th s yth wa e sar l significancf eo detections of (or upper limits to) non-gaussianity. CMB Photons are last scattered by electrons at the recombination epoch. ItSs a Thomson scattering. If the distribution of incoming radiation has a quadrupole moment, the scattered radiation has some degree of linear polarization. The degree of linear anisotropye polarizatioth f o ordee % th expecte e f £f r10 o :th o s ni d signae thus lth i f so lesr o ss jtK(sew rm fe ordee a e.g f ro . [72]). Despit lonea g lasting experimental effort (see e.g. [73, 74, 75, 76, 77, 78]) the polarization of the CMB has not been detected bese yetTh .t uppe 10^x ordee r6 AT/T n th i limit6f f o t rangula o a dat o t se ear r scales around one degree. In the standard scenario [79] we expect acoustic peaks in the polarization power spectrum at I £ 200. The presence of tensor perturbations generated by inflation produces a curl component B in the CMB polarization field which adds to the curl-free E component described above. Measuring the power spectrum of polarization of the CMB is very important for several reasons. First, it provides four power spectra to measure: in addition to the < TT > power spectrum of the anisotropy T, it is possible to measure the < TE > anisotropy-polarization cross-spectrum; the pure polarization powe mors ri spectru> e challengingE E m< componen> B B , whil< evee s i t eth n smalle non-zers i d ran o onl gravity-waveyf i s were presen last ta t scattering detectios .It n would prove directly the physics of inflation. The < EB > and < TB > spectra are zero by parity. The BOOMERanG instrument has been modified including a new focal plane with polarization senitive bolometers: its new long duration flight (called B2K) is

169

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 2001f o forecase scheduled .en Th e th t r sensitivitdfo polarizatioyo t bees nha n computed in [80]. Many other attempts to measure CMB polarization are in progress( see e.g. [81] and references therein), and a first detection is expected soon .

ACKNOWLEDGMENTS

The BOOMERANG project has been supported by the CIAR and NSERC in Canada, by Programma Nazionale Ricerche in Antartide, Universita "La Sapienza", and Agenzia Spaziale Italiana in Italy, by PPARC in the UK, and by NASA, NSF OPP and NERSC in U.Se receiveth e .W d superb fielflighd dan t support from USAe NSBth d PF an personne l in McMurdo.

REFERENCES

1. Peebles, P.J.E, and Yu J.T., 1970, Ap.J. 162, 815 2. Sunyaev, R.A. & Zeldovich, Ya.B., 1970, Astrophysics and Space Science 7, 3 3. Doroshkevich A.G., Zeldovich Ya.B., Sunyaev R.A., 1978, Soviet Astronomy, 22, 523 4. Silk J. & Wilson M. L. 1980, Physica Scripta, 21, 708 Sugiyam, W. u Sil & , H 1997 k. J. a. N 5 , Nature, 3867 ,3 6. Martinez-Gonzalez E., Sanz J.L. & Silk J., 1990, ApJ, 355, L5 7. Weinberg S., Rev. Mod. Phys. 61 1 1989 8. Weinber . astro-ph/000627gS 6 Efstathio. . BondG R . . J 9 ,d Monuan . Not Astron. R . . Soc. 304 (1999)5 ,7 . . Melchiorr10 Griffithd an . iA s L.M., astro-ph/0011147 11. Seljak, U. & Zaldarriaga, M. 1996, ApJ. 469,437 . Lewi12 , StewarsA. Lasenb, tE. , astro-ph/991117yA. 6 . Bucher13 Moodley, ,M. Turok& . 2000. ,K ,N , astro-ph/0007360 14. Smoot G.F. et al, 1992, ApJ, 396, LI 15. Bennett C.L. et al., 1996, ApJ, 464, LI . Gut16 h A.H., 1982, Phil. Trans . Soc..R , A307,141 . Lind17 , 1982eA. , Phys.Lett, 108B9 38 , 18. Linde A., 1983, Phys.Lett, 129B, 177 19. Albrecth A. & Steinhardt P.J., 1982, Phys.Rev.Lett, 48,1220 . Gut20 h A.H. 1997 Inflationare Th , y Universe theorw ne ques cosmif e ya o r :Th tfo c originis, Addison- Wesley Yorkw ,Ne . 21. Kolb E.B. & Turner M., 1990,'The Early Universe', Addison-Wesley, New York 22. Kinney W, Melchiorri A., Riotto A., PRD in press, astro-ph/0007375 (2000) 23. Vilenkin A. & Shellard E.P.S., 1994, 'Cosmic Strings and other Topological Defects', Cambridge University Press . Bond Quintessentiaal.t e 24 e , ,Th .R. l CMB, PasFutured tan , CAPP200Proce th f .o 0 meeting, Verbier, astro-ph/0011379 25. J. R. Bond, G. Efstathiou and M. Tegmark, Mon. Not. R. Astron. Soc. 291, L33 (1997) and references therein 26. Pospieszalski, M., IEEE-MTT-S Digest, 1369, (1992) 27. Pospieszalski, M., IEEE-MTT-S Digest, 1121, (1995) . Torbe al.t 28 e , ApJ.tE. , 521, L79-L8 2(1999, ) 29. Miller A., et al., ApJ., 524, L1-L4 , (1999) 30. Mauskopf P., et al., Applied Optics, 36, 765-771, (1997) . Mauskop31 al.t e , , ApJ.fP. , 536, L59-L62 (2000) Bernardie d . al.t e 32 , , NatursP. e 404,955 (2000).

170

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp 33. Hanan alt e ,, Ap.J.yS. , 545, L5-L9 (2000) 34. Delabrouille J. et al., MNRAS, 298,445-450 (2000) Bernardie d . Mas35 & , . 1998sP S. i 'Fundamentan ,i l parameter cosmology'n si , XXXIIIre Procth f .o d rencontre Morione sd d (France), Iran , Giraoud-HerauThanJ. n hVa Bouche, dY. , DamouF. t & . rT Mellie eds.. rY , Editions Frontieres, p.209 36. Piacentini F., et al., 2001, astro-ph/0105148 al.t . Crile , 37 2001lB. preparationn i , . cosmology. BorrilK 3 38 n i . lJ , Rom 476,277aP 1998C P ,AI , (1999). . Prunet39 al.t e ,. 2000S , "Energn ,i y densitie Universe"e th n si , Bartlet , DumarchetJ. . eds.zJ , Editions Frontieres, Pari sastro-ph/000605- 2 Bernardise d . P . t al.40 e , , §First results fro BOOMERane mth G experimen CAPP200 e Proc, tf th f o . 0 meeting, Verbier, July 2000, astro-ph/00011469 2001. Letteral J . Mast e 41 Ap ,, pressn siS. i , astro-ph/0101539 . WOMBA42 T collaboration, 1998 http://astron.berkeley.edu/wombat/foregrounds/radio.htmle se , . 43. Hivon E. et al., astro-ph/0105302 44. Padin et al., 2001, astro-ph/0012211 45. Dawson K.S. et al., 2000, astro-ph/0012151 . http://astro.uchicago.edu/dasi46 / 47. http://astro.caltech.edu/ tjp/CBI/ 48. http://www.mrao.cam.ac.uk/telescopes/vsa/index.html 49. http://www.archeops.org/ 50. http://www.topweb.gsfc.nasa.gov/ Bernardise d . Troiae P D . . . Migli51 L G ,, o "Calibratio f balloon-bornno experimentsB eCM W "NE ASTRONOMY REVIEWS, 43, 281-287,1999. 52. Jaffe, A., et al., 2001, PRL, 86, 3475-3479 . Kno53 Pag, x L. Phys.Rev.Lett, eL. (20005 8 . ) 1366-1369 54. Donahue M., Voit M, astro-ph/9907333 55. Bahcall N.A. et al, ApJ. 541,1 (2000) 56. Blakesee J.P., et al., astro-ph/9910340 57. Juszkiewicz R., et al., Science 287,109 (2000) 58. Wittman et al., Nature, May 11 2000 . Ostrike59 r J.P.Steinhardd ,an t P.J., Nature, 377(19950 60 , ) 60. Caldwell R.R., Dave R., Steinhardt P.J., Phys. Rev. Lett., 80, 1582 (1998) . Armendari61 Mukhano, zC. , SteinhardvV t P.J., astro-ph/0004134 (2000) 62. AmendolaL.,astro-ph/0006300 63. Riess A.G. et al, ApJ. 116,1009 (1998) 64. Perhnutter S. et al, ApJ. 517, 565 (1999) 65. Linde A.D., PhysLett. 108B, 389, (1981) . Watson66 , G.S., astro-ph/0005003, (2000) . 2000 . Tytleral t 67 e ,. Physic,D a Scripta submitted, astro-ph/0001318 . Buries68 , NollettS. , , K.M Turner.& , M.S. 2000, astro-ph/0010171 69. Gawiser E., astro-ph/0105010 (in this book) Barreiro. B . R . , 70 astro-ph/9907094 71. J.H.P.Wu et al., 2001, Tests for Gaussianity of the MAXIMA-1 CMB Map , astro-ph/0104248 72. Kaise , 1983rN. , M.N.R.A.S., 101,1169 73. Caderni N., et al., 1978, PRD, 17. 74. Lubin P., Smoot G., Ap.J., 245,1 75. Lubin P., etal, Ap.J., 273, L51 . Wollac76 al.t ke , 1993, Ap.J., 4199 ,L4 . Netterfiel77 al.t e , 1996dB. , Ap.J., 445, L69. 78. Pisano G., New Astronomy Reviews, 2000, 43, 329-339 79. A. Melchiorri, N. Vittorio, astro-ph/9610029 80. TegmarkM. et al. Astrophys.J. 530 (2000) 133-165 . Staggs81 al.t e ,. astro-ph/990406S , 2

171

Downloaded 02 Oct 2007 to 131.215.225.176. Redistribution subject to AIP license or copyright, see http://proceedings.aip.org/proceedings/cpcr.jsp