Absolute measurements of the cosmic microwave backgroundfrom Amundsen-Scott South Pole Station

M. BERSANELLI, G. BONELLI, and G. SIRONI, Istituto di Fisica Cosmica, Consiglio Nazionale delle Richercheand Universitd degli Studi, Milan, Italy

S. LEVIN'Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109

G.F. SMOOT, M. BENSAD0uN, G. DE Amid, M. LIMON, and W. VINJE*, Lawrence Berkeley Laboratory andSpace Science Laboratory, University of California, Berkeley, California 94720

'Present address: Physics Department, Princeton University, Princeton, New Jersey 08544.

Observations of the cosmic microwave background (CMB)play a central role in modern cosmology. The existence

of the CMB as a remnant of the early Universe was predictedbefore its first detection, and its discovery (Penzias and Wil-son 1965) constituted a pillar for the Big Bang cosmologicalscenario against rival theories. The following 2 decades led tostrong experimental confirmation of the main characteristicsof the CMB radiation field expected from the Big Bang theory:high degree of isotropy, very low polarization level, and ther-mal (blackbody) spectrum at low [approximately 2.7 Kelvin(K)] temperature. The recent cosmic background explorer(COBE) differential microwave radiometer (DMR) resultshave provided further support to the generally accepted stan-dard model by detecting for the first time primordial fluctua-tions in the CMB field at the limit expected by structure for-mation theories (Smoot et al. 1992).

Energy injections to the CMB photons related to forma-tion of large-scale structures or to other physical processesoccurring up to redshifts z:5106 may have left signatures in theCMB as distortions from a purely Planckian spectrum. COBEfar infrared absolute spectrophotometer (FIRAS) has tremen-dously improved the accuracy of the determination of theCMB spectrum over previous experiments above 30 gigahertz(GHz) [wavelengths <1 centimeter (cm)] and found no evi-dence of spectral distortions at the 1 percent level (Mather etal. in press). At lower frequencies, however, where spectraldistortions are expected to be largest, ground-based measure-ments still provide the best observational limits. Since 1982we have been involved in an international program ofground-based absolute measurements of the CMB at cen-timeter and multicentimeter wavelengths.

The need for high accuracy in the results and improvedinstrumental sensitivity have pushed us, as all CMBresearchers, to base the experiments in low-background envi-ronments. Balloon- or satellite-based measurements areexcellent in principle, but they are not feasible for our long-wavelength absolute measurements that require large, heavyantennas and calibrators. In the first phase of our collabora-tion (1982-1988), we measured from the White MountainResearch Station, California, whereas in 1989 and 1991, weobserved from the South Pole.

The principle of the measurement (see, for example,Smoot et al. 1987, 1991) is to determine the intensity of thezenith sky by comparison with the signal from a calibrator

cooled to the boiling temperature of liquid helium; all thelocal emissions (from the atmosphere, the ground, theGalaxy, for example) contributing to the sky signal are thenmeasured and subtracted to obtain the CMB intensity as aresidual. The instruments are either total-power or differen-tial Dicke radiometers and are calibrated with different tech-niques (see, for example, Kogut et al. 1991).

The choice of the site is intended to minimize the uncer-tainties related to each correction term. The surroundings ofthe Amundsen-Scott South Pole Station, or of possible futurestations in the antarctic plateau, are most likely the best pos-sible choices for Earth-based absolute CMB measurements.The polar environment turns out to be a unique locationwhere a number of peculiar requirements, all critical for thesemeasurements, coexist (table 1).

In our first antarctic expedition (November and Decem-ber 1989), we studied the CMB spectrum by measuring at0.82, 1.47, 2.5, 3.8, and 7.5 GHz (see table 2) and found resultsin very good agreement with our measurements at the sameor similar frequencies performed from White Mountain. Thisagreement is an indication that site-dependent systematiceffects (ground, atmosphere, radio frequency interferences)are likely to be well understood and properly accounted for.In particular, we confirmed our result at 1.47 GHz to beapproximately 2.5 standard deviation lower than the globalaverage CMB temperature of 2.73 K (Bensadoun et al. 1993).

Table 1. Properties of the South Pole site

High altitude2,830 mAtmospheric emissionFlat horizon-900 everywhereGround emissionDry atmosphere<1 mm H20Atmospheric emissionHighly isolated Radio frequency

interferencesLogistic supportOutstandingResearchers, experimentDust free Antennas,

Sun-ground screensShapable surface Experiment layoutbCold ground-40°CGround emissionLow winds<15 knotsEase of operationaAustral summer.b5ee the figure.

ANTARCTIC JOURNAL - REVIEW 1993306

LII-

In 1991, we returned to the South Pole with amodified version of the 1.47-GHz radiometer andTable.

improvements in the liquid-helium calibrator; wealso built a new independent radiometer operat-ing at a nearby frequency (2.0 GHz) to further =crosscheck this result (figure). Table 2 summa-rizes the radiometers operating in our two antarc-0.82tic campaigns and the obtained values of the CMBthermodynamic temperature, TCMB. Our prelimi-1.47nary analysis at 2 GHz gives TCMB=2.55±0.14 K(Bersanelli et al. in press). This result is consistent,

2.0approximately 1 standard deviation, with anunperturbed Planckian spectrum, but it does not2.5rule out the lower measured value of TCMB foundat the lower frequency. The analysis of the 1.473.8GHz 1991 measurement is underway. The presentresults limit any possible energy release in theearly Universe (from 1 to 100,000 years after theBig Bang) to less than 0.1 percent of the total CMBaAnalysienergy.

The accuracy of our measurements at frequencies belowapproximately 2.5 GHz is limited by the subtraction of galac-tic emission due to synchrotron and free-free radiation, aforeground signal that cannot be eliminated by the choice ofobserving site. We model the galactic contribution based onlow-frequency maps scaled at the frequency and angular res-olution of our instruments and test the model assumptionswith differential scans. We are presently involved in a long-term effort (De Amici et al., Antarctic Journal, in this issue) toimprove our knowledge of the galactic emission and spectralindex, a step necessary for future improvement of low-fre-quency measurements of the CMB spectrum.

This work has been supported by National Science Foun-dation grant OPP 90-18395, by the U.S. Department of Energy

. Summary of the South Pole measurements

36.619892.7±1.6Sironi, Bonelli, andLimon 1991

20.419892.26±0.20Bensadoun et al. 19931991a

15.019912.55±0.14Bersanelli et al. 1993

12.019892.50±0.34Sironi et al. 19911991a

7.919892,64±0.07De Amici et al. 1991

4.019892.69±0.07Levin etal. 19921991a

s in progress.

under contract DE-ACO3-76SF00098, and by the ItalianAntarctic Program. We wish to thank Francesco Cavaliere,John Gibson, Andrea Passerini, and John Yamada for techni-cal assistance.

ReferencesBensadoun, M., M. Bersanelli, G. De Amici, A. Kogut, S. Levin, M.

Limon, G.F. Smoot, and C. Witebsky. 1993. Measurements of thecosmic microwave background temperature at 1.47 GHz. Astro-physical Journal, 409, 1.

Bersanelli, M., M. Bensadoun, G. De Amici, M. Limon, S. Levin, G.F.Smoot, and W. Vinje. In press. Absolute measurement of the cos-mic microwave background at 2 GHz. Astrophysical Journal.

De Amici, M., M. Bersanelli, A. Kogut, S. Levin, M. Limon, and G.F.Smoot. 1991. The temperature of the cosmic background radiation

The atmospheric emission is mea-sured with scans at varying anten-na angles from the zenith wherethe signal is correlated to the air-mass. This picture shows simulta-neous atmospheric measurementswith the 1.47 GHz (left) and 2.0GHz (right) radiometers. Theinstruments are based at the bot-tom of an approximately 4-meter-deep pit excavated in the ice.Large aluminum shields surroundthe instruments to protect theobservations from unwantedemission from the ground and theSun. From the same location weperformed differential measure-ments of the galactic emission. Asimilar site set-up was realized forthe absolute calibration measure-ments.

ANTARCTIC JOURNAL - REVIEW 1993307

at 3.8 GHz: results of a measurement from the South Pole site.Astrophysical Journal, 381, 341.

De Amici, G., G.F. Smoot, M. Bensadoun, M. Limon, W. Vinje, M.Bersanelli, G. Bonelli, and G. Sironi. 1993. Low-frequency maps ofthe galactic radio emission. Antarctic Journal of the U.S., 28(5).

Kogut, A., M. Bensadoun, M. Bersanelli, G. De Amici, S. Levin, M.Limon, G.F. Smoot. 1991. A decade of long-wavelength CMBmeasurements. In C.L. Bennett, V. Trimble, and S. Holt (Eds.),After the First Three Minutes Workshop, AlP Conference Proceed-ings 222, University of Maryland, College Park, Maryland. NewYork: AlP.

Levin, S., M. Bensadoun, M. Bersanelli, G. De Amici, A. Kogut, M.Limon, and G.F. Smoot. 1992. A measurement of the cosmicmicrowave background temperature at 7.5 GHz. AstrophysicalJournal, 396, 3.

Mather, J.C., E.S. Cheng, D.A. Cottingham, R.E. Eplee, Jr., D.J. Fixsen,T. Hewagama, R.B. Isaacman, K.A. Jensen, S.S. Meyer, P.D.Noerdlinger, S.M. Read, L.P. Rosen, R.A. Shafer, E.L. Wright, C.L.Bennett, N.W. Boggess, M.G. Hauser, T. Kelsall, S.H. Moseley, Jr.,R.F. Silverberg, G.F. Smoot, R. Weiss, and D.T. Wilkinson. In press.Measurement of the cosmic microwave background spectrum bythe COBE FIRAS instrument. Astrophysical Journal.

Penzias, A.A., and R.W. Wilson. 1965. A measurement of excess anten-na temperature at 4080 Mc/s. Astrophysical Journal, 142, 419.

Sironi, G., G. Bonelli, and M. Limon. 1991. The brightness tempera-ture of the South Celestial Pole and the temperature of the cosmicbackground radiation measured at 36.6 and 12 centimeter wave-length. Astrophysical Journal, 378, 550.

Smoot, G.F., C.L. Bennett, A. Kogut, E.L. Wright, J. Aymon, N.W.Boggess, E.S. Cheng, G. De Amici, S. Gulkis, M.G. Hauser, G. Hin-shaw, C. Lineweaver, K. Loewenstein, P.D. Jackson, M. Janssen, E.Kaita, T. Kelsall, P. Keegstra, P. Lubin, J.C. Mather, S.S. Meyer, S.H.Moseley, T.L. Murdock, L. Rokke, R.F. Silverberg, L. Tenorio, R.Weiss, and D.T. Wilkerson. 1992. Structure in the COBE differen-tial microwave radiometer first-year maps. Astrophysical JournalLetters, 396, Li.

Smoot, G.F., M. Bensadoun, M. Bersanelli, G. De Amid, A. Kogut, S.Levin, and C. Witebsky. 1987. Long-wavelength measurements ofthe cosmic microwave background radiation spectrum. Astrophys-ical Journal Letters, 317, L45.

Smoot, G.F., G. De Amici, M. Bensadoun, A. Kogut, S. Levin, M.Limon, G. Sironi, M. Bersanelli, and G. Bonelli. 1991. The long-wavelength spectrum of the cosmic microwave background.Antarctic Journal of the U.S., 26(5), 286.

Low-frequency maps of the galactic radio emissionG. DE Amid, G. SMOOT, M. BENSADOUN, M. LIMON, and W. VINJE*, Lawrence Berkeley Laboratory and

Space Science Laboratory, University of California, Berkeley, California 94720M. BERSANELLI, G. BONELLI, and G. SIR0NI, Istituto di Fisica Cosmica, Consiglio Nazionale Ricerche and

Universitd degli Studi, Milan, Italy

*Present address: Physics Department, Princeton University, Princeton, New Jersey 08544.

The field of radio astronomy started, almost by chance,when C. Jansky, studying short-wave interferences to

radio broadcasts, discovered radio emission coming from thedirection of the galactic center. The year was 1932. As thetools and techniques have improved, radio astronomy hasprovided scientists with opportunities to probe a multitude ofcosmic phenomena with unprecedented resolution, leadingto the discovery of unexpected facets of the Universe, whilesteering the efforts away from the determination of the galac-tic emission and toward studies at small angular scales of dis-tant objects. Among the most important discoveries of radioastronomy is the cosmic microwave background (CMB) radia-tion; recent improvements in its studies (for example, Smootet al. 1992; Bensadoun et al. 1993) have made a precise deter-mination of the diffuse foreground galactic signal importantagain.

The galactic emission is composed of the superpositionof the signals generated by acceleration of relativistic elec-trons in the galactic magnetic field (synchrotron radiation),by thermal bremsstrahlung inside hydrogen clouds (Hii orfree-free radiation), and by thermal radiation of dust clouds(dust radiation). The study of galactic emission brings a betterunderstanding not only of the morphology but also of the

energy balance of the different components and of the distri-bution of magnetic fields and free electrons.

Because the CMB radiation is supposed to be the leftoverfrom a hot, dense phase of the early Universe, in the CMB weexpect to find the fossil remnants of the processes that haveshaped the Universe into its present form. The shape of thefrequency spectrum of the CMB is well known above 60 giga-hertz (GHz) (Mather et al. in preparation), but much less pre-cisely determined at the low-frequency end of the spectrum(see, for example, Bersanelli et al., Antarctic Journal, in thisissue), where the strong foreground galactic signal cannot bereduced by either instrument design or observational tech-niques. Similarly, measurements of anisotropy rely on accu-rate subtraction of the ubiquitous (and irregularly distrib-uted) foreground emission. Only a handful of large-area sur-veys of the sky emission are available in the literature, andonly one of them covers the whole sky; on this rather sparsesample, which is also affected by inconsistent zero levels anduncontrolled gain changes, rests the accuracy of all spectrumand anisotropy measurements in the radio and microwaveregion.

Our group at Berkeley has started a new program, thegalactic emission mapping (GEM) project, to improve the pre-

ANTARCTIC JOURNAL - REVIEW 1993

308