Nobel Lecture: Plurality of worlds in the cosmos: A dream of antiquity, a modern reality of astrophysics^*

Figure 1

200 billion stars, but how many planetary systems are there in the Milky Way? This photo illustrates the huge number of stars seen in a very small fraction of the disk of our Galaxy. How do we detect planetary systems hosted by these stars?

Figure 2

1995, images of protoplanetary disks: The Hubble Space Telescope reveals protoplanetary disks around very young stars of the Orion Nebula. From [32].

Figure 3

Alex Wolszczan and Michel Mayor in front of the secondary mirror of the Arecibo radiotelescope in 2002 (twenty year celebration of the discovery of the planetary system hosted by the neutron stars PSR B1257+12).

Figure 4

The CORAVEL template used to determine the radial velocity of states by cross-correlation 40 years ago! To be seen at the Nobel Museum. From [1].

Figure 5

Roger Griffin was the first to demonstrate, already in the early sixties, that radial-velocity measurements can be efficiently made by a cross-correlation spectrograph ([18]).

Figure 6

Spectrum of HD 85512 obtained with the ESPRESSO spectrograph installed at the ESO Paranal Observatory (Chile). Only $400\hspace{0.17em}\hspace{0.17em}\AA$ appear in the figure while the true spectral window of ESPRESSO is about 10 times larger. With ESPRESSO we can measure the important Doppler information contained in the spectra of solar-type stars or colder. The cross-correlation technique allows one to concentrate Doppler information from several thousand absorption lines to measure precise stellar radial velocities.

Figure 7

André Baranne, the father of the optics of CORAVEL, ELODIE and CORALIE.

Figure 8

The ELODIE instrument installed on the 1.93-meter telescope at the Haute-Provence Observatory (OHP) was built by the technical staffs from OHP and Geneva Observatory. A special mention is due to André Baranne (Optical Engineer at Marseille Observatory). André is in the second row, just below the word OHP. He discovered the white pupil mounting, broadly used today in many astronomical spectrographs. (Alain Vin is missing from this photo!)

Figure 9

Antoine Duquennoy.

Figure 10

Radial velocity measurements of 51 Pegasi obtained during four observing runs from September 1994 to September 1995. Adapted from [29].

Figure 11

The first exoplanet hosted by a solar-type star: 51 Pegasi b (October 1995). From [29].

Figure 12

Exoplanet pioneers at the Wyoming conference 2011. From left: Alex Wolczscan, Michel Mayor, Nathalie Bathalia, William Borucki, David Charbonneau and Geoff Marcy.

Figure 13

The observed diversity of planetary systems shows (i) orbital periods as short as a few hours and (ii) orbits with a large range of eccentricities. Period-eccentricity diagram for the sample of known exoplanets (in 2007!) in comparison with stellar binaries. The Earth and giant planets of the Solar System are indicated as well. From [40].

Figure 14

M. Mayor and D. Queloz at La Silla Observatory (ESO, Chile) in front of the 1.2-meter EULER telescope and in the distance, the 3.6-meter telescope. These two telescopes have made significant contributions to the detections of exoplanets since 1998 and 2003, respectively.

Figure 15

September 9 and 16, 1999. A first planetary transit. Hot Jupiters are gaseous giant planets: $\mathrm{d}\mathrm{e}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{t}\mathrm{y}=0.3\mathrm{g}/{\text{cm}}^3$. (a) From [11]. (b) From [7].

Figure 16

Metallicity distribution of planet hosting stars ([28]). Left panel: the frequency of giant planets as a function of stellar metallicity is shown based on results from the HARPS planet search program. In the right panel, we illustrate the same plot for stars that host only Neptune- or Super-Earth-like planets. These plots shows a clear correlation between the presence of giant planets and the metallicity of the star. This trend is not seen for stars hosting a lower-mass planet.

Figure 17

Garik Israelian and Nuno Santos: two colleagues having contributed so much to the study of the chemical composition of stars with or without planets.

Figure 18

After 16 years we continue to conduct a large and systematic survey of stars in the southern hemisphere with the CORALIE spectrograph on the EULER telescope and the HARPS spectrograph on the 3.6-meter telescope. We try to design surveys with controlled detection bias in order to obtain distributions of planetary systems as functions of mass, orbital period, host star metallicity, etc., These distributions provide constraints on planet formation scenarios and tell us (for example) that planets more massive than 50 Earth-masses are hosted by about 14% of solar-type stars. We also remark that “hot-Jupiters,” while being the first planets detected are rare (about 1%), most giant planets have larger orbital periods from several months to several years. We can also remark on the extreme abundance of planets with masses between a few Earth-masses and 20 Earth-masses (Super-Earth).

Figure 19

Since 1995: A huge number of discoveries and improvement of astronomical instrumentation. After the discovery of 51 Pegasi b, we have observed an amazing number of planet discoveries. We can also remark on the result of the improvement of the sensitivity of Doppler spectrographs allowing today the detection of planets with masses as small as the Earth mass (at least for relatively tight orbits!).

Figure 20

Mass-radius diagram of planets smaller than 2.8 Earth radii ([16]). The dashed lines show planetary interior models for different compositions as labeled ([44]). Planets are color-coded according to the incident flux $F_p$, relative to the solar constant received on the Earth [for the full description see 16].

Figure 21

Increasing the precision. Radial velocity via cross-correlation spectroscopy: A path to the detection of Earth-type planets. Over the last 40 years the precision for the different generations of cross-correlation spectrographs has been increased by a factor 3000! This gain of sensitivity allows for the discovery of planets with smaller masses. Francesco Pepe (Fig. 22) is the PI of the team having built ESPRESSO.

Figure 22

Francesco Pepe, the principal investigator of the ESPRESSO spectrograph. This instrument, installed at Cerro Paranal (ESO Chile) can feed by one either one or four 8.2 meter unit telescopes of the VLT, achieving thus a collecting power equivalent to a 16-meter telescope! ESPRESSO represents the latest generation of the series of our cross-correlation spectrographs and was designed to achieve a precision of $0.1\mathrm{m}/\mathrm{s}$.