A.2 - Scientifics goals

The characteristics of the angular fluctuations of the 2.728 K Cosmic Microwave Background (CMB) are a goldmine of cosmological information. They tell us about the state of the Universe (at a redshift of typically 1000) when it was only a few hundred thousand years old. During this epoch, the matter fluctuations, which are bound to develop by the gravitational instability into galaxies, clusters of galaxies and the large-scale structures of the Universe, have a very weak amplitude (few parts per million). The prediction of the their characteristics within a cosmological model is simplified because the computation of their evolution can be linearised.

These density fluctuations generate, at the epoch of the decoupling between matter and radiation, fluctuations of temperatures of the CMB that can be observed today as anisotropies. There is therefore a rather direct link between the distribution of energy in the fluctuations at various scales and global information (that cannot be reached by other means) such as the primordial anisotropy spectrum and the cosmological parameters describing the state of the Universe at the moment when the anisotropies were frozen.

Figure 1 shows an example of the the dependance of the primordial anisotropy spectrum on the value of some cosmological parameters. Hence, accurate measurements can yield very strong constraints on the ten fundamental parameters or so that cosmologists have been trying to constrain with "local" measurements. The tracers of the local (hence present) Universe, mostly the galaxies, are only the remote end of initial conditions, transformed in a non-linear fashion by nearly 15 billion years of evolution.

Fig. 1: Example of the dependance of the primordial anisotropy spectrum on the value of some cosmological parameters. Abscissae show the spherical harmonics order (corresponding to a typical angular scale shown above). Ordinate scale shows the temperature fluctuation rms power in arbitrary units (Courtesy of Caltech).

The fundamental parameters that can be constrained by the anisotropy measurements are mainly the Universe expansion rate i.e. the Hubble Constant, the energy density in the form of baryons, dark matter, vacuum (i.e. the cosmological constant) and gravitational waves, as well as the ionisation history of the Universe. It is also possible to constrain the parameters that are necessary to the description of the spectrum of primordial fluctuations that were generated at the very beginning of the Universe, that give information on physics at energies that are beyond particle physics standard model and above the limit provided by large particle accelerators.

The goal of a large international community is therefore the measurement with as good an accuracy as possible of the characteristics of the spatial fluctuations of the CMB. Since the first detection of these fluctuations by the COBE satellite of the NASA, numerous ground-based and balloon-borne projects are aiming toward that goal. On a longer term, 2 space projects are decided: MAP must be launched by NASA in 2002 and the ESA mission PLANCK will yield a definite map of the primordial anisotropies starting in 2007. The Planck mission is both the most ambitious (in particular the bolometer HFI instrument) and the most remote in time.

The aim of the ARCHEOPS balloon project is twofold:

- It will allow the consortium to have access to high quality CMB data with an unmatched sky coverage. It would yield, before MAP, competitive results on the CMB anisotropy measurements;

- It will be a testbed with real data for the many technics of data analysis that are planned for PLANCK.

The comparison with the current status of observations (Fig. 2) shows the usefulness of ARCHEOPS-like data. By applying to ARCHEOPS the methods that were devised to determine the scientific return of PLANCK, one can perform a first estimate of the errors that can be expected on the anisotropy spectrum (Fig. 3), once the contaminating foregrounds have been dealt with. Hence the results are obtained after a realistic foreground estimate and removal are applied to simulated data. Indeed it is crucial to separate additional components of the (sub)millimetre sky from the CMB. Moreover, ARCHEOPS constitutes a critical step in the preparation of some of the HFI community to succeed in a short time (PLANCK proprietary period of two years) to reduce and exploit the data which will be difficult to analyse specially during the checks of systematic effects.

ARCHEOPS data will actually be first non-simulated data on which the algorithms that are developped today can be tested and improved.

Beyond the scientific output of the balloon, the data will help forming the PLANCK users community. ARCHEOPS is an important step in the long-term strategy of maximal return for our community from the heavy investments.

Fig. 2: Observational status of anisotropy measurements as a function of the angular scale (l is the spherical harmonic order in absissae, l(l+1)Cl is the power spectrum of the spatial fluctuations in ordinate). The various points and error bars show the present results that were obtained in different experiments (ground-based, balloon and satellites). The solid curve is the CDM model prediction with the canonical parameters: H0 = 50 km / s / Mpc , W0=1, and Wb = 0.05.

Fig. 3: This figure shows the accuracy with which the power spectrum of anisotropy can be determined with ARCHEOPS. The theoretical spectrum in black corresponds to the standard model with H0 = 50 km / s / Mpc , 0=1, and b = 0.05. The two curves in red show the error bar that can be expected with one ARCHEOPS flitght. The corresponding accuracy on the cosmological parameters H0, 0 , and b is better than 10%, depending on the models. The error bar computation was obtained by using a realistic estimate of the foregrounds (interstellar dust, free-free, synchrotron) and applying their removal to simulated data. This figure shows the end result of the sky simulation and inversion procedure (using the 3 spectral bands) that were set up for the PLANCK HFI proposal (Gispert & Bouchet 1998) and adapted to the ARCHEOPS case (arctic flight and 25% sky coverage).

An estimate of the anisotropy power spectrum can be obtained from these maps. For given detector performances, the estimate will improve with the sky coverage and angular resolution as well. The use of cold bolometers at the focus of a 1.50m (1.30 effective) telescope allows us to reach an angular resolution better than 10 arcmin at millimetre wavelengths. The goal for ARCHEOPS is thus to map 25% of the sky with 10' resolution and a sensitivity per sky pixel of 20' of T/T ~ 3.10-5. The solution that satisfies these requirements is a 24 hr balloon flight (a complet rotation of the Earth) at the Esrange CNES station (on the Arctic circle) during the Winter to avoid the Sun. The telescope points at 45 deg from zenith. It spins around the vertical axis of the gondola pivot at a typical rate of 2 rpm, the diurnal Earth rotation producing the total coverage of a sky annulus. The redundancy is of half a beam after one period. Figure 4 shows the sky coverage obtained during the flights from Esrange (Kiruna, Sweden) and during the test flight in Trapani (Sicily). Figure 5 shows a simulated sky as observed by ARCHEOPS in Kiruna, compared to the "true" sky.

Fig. 4: Sky coverqge for the flights in Trapani and at the Esrange Station. The covered zone limits are superposed to a map of galactic emission as anticipated at 150 GHz, extrapolated in frequency from DIRBE dust emission map at 240 microns, and from the synchrotron emission map at 408 MHz by Haslam C.G.T., Quigley M.J.S. et Salter C.J. (1970, MNRAS 147, 405). The map is a sine projection in equatorial coordinates. The region covered by the 6hr Trapani flight will contain a small part of the centre of the Galaxy (right) and a large part of the Galaxy in the anti-centre (left).

The region covered by the 24 hr flight at the Esrange station is an annulus made by the declinations between 22 and 68 degree. It contains regions where the dust galactic emission is minimal (at the centre of the map) where CMB anisotropy will be the least confused.

Fig. 5: Top panel: simulated sky at 143 GHz (2.1 mm) in galactic coordinates. The sky at this frequency is dominated by the CMB dipole and the galactic emission. The observed positions is such, across the cosmological dipole, that the dipole can be used for calibrating the signal, scan by scan. Dark regions show the area observed during the Kiruna flight. Bottom left panel shows data simulated with a realistic noise projected onto the sky, after a simplified data reduction has been applied (deglitch, calibration, dipole subtraction). Bottom right panel shows the "true" sky i.e. without instrumental noise nor dipole. Comparison of the 2 bottom maps show that the CMB is apparent in the ARCHEOPS simulation for galactic latitudes larger than 30 degrees. Stripping effects can be seen near the Galactic plane because of bright sources.