All Figures

$\begin{figure} \includegraphics[width=8.5cm,clip]{6194fig1.eps}\par \end{figure}$ Figure 1: Phase-folded radial-velocity measurements (in km s^-1) and best fit for the two planets in the two-planet solution. For every planet the contribution by the other planets has been subtracted. In the upper panel showing the long-period planet, a time-correlated structure becomes evident. HARPS data are plotted in red, UCLES data in blue, and CORALIE data in green.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.4cm,clip]{6194fig2.eps}\par \end{figure}$ Figure 2: Phase-folded radial-velocity measurements (in km s^-1) and best fit for any of the four planets. For every planet the contribution by the remaining three planets has been subtracted. HARPS data are plotted in red, UCLES data in blue, and CORALIE data in green. Note that on the third panel showing the longest-period planet the HARPS data clearly indicate a flattening of the radial velocity curve. The recent HARPS data consequently constrain the orbital period of this planet.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[angle=270,width=8.8cm,clip]{6194fig3.ps}\par \end{figure}$ Figure 3: Global solution containing 4 planets and covering the full time range of observations. The lower plot shows the residuals of the fitted function to the actual measurements. The fit takes planet-planet interaction into account. The weighted dispersion of the residuals is 1.75 m s^-1.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.3cm,clip]{6194fig4.eps}\par \end{figure}$ Figure 4: The figure shows a zoom on the time span covered by the HARPS data. It must be note that HARPS data constrain in an exceptional way the three shorter-period orbit. It particular, these complex curve would be incompatible even with a two-planet solution.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8cm,clip]{6194fig5.eps}\par \end{figure}$ Figure 5: The upper panel shows the phase-folded HARPS data and the best Keplerian best fit to the P=9.64 days-planet after subtracting the contribution of the other planets. The lower panels shows the residuals of the data to the best fit. The dispersion of the HARPS residuals is 1.41 m s^-1 rms.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.8cm,clip]{6194fig6.ps}\par \end{figure}$ Figure 6: Orbital evolution of the four planets over one million years starting with the orbital solution from Table 1 and an inclination of $90^\circ$ . The panel shows a face-on view of the system; x and y are spatial coordinates in a frame centered on the star. Present orbital solutions are traced with solid lines and each dot corresponds to the position of the planet every 50 years. The semi-major axes are constant, and the eccentricities undergo small variations ( 0.09 < e_b < 0.13; 0.16 < e_c < 0.21; 0 < e_d < 0.19; 0.08 < e_(e) < 0.11).
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.8cm,clip]{6194fig7.ps}\par \end{figure}$ Figure 7: Global view of the dynamics of the $\mu$ Arae system for variations in the eccentricity with the semi-major axis of $\mu$ Arae d. The color scale is the stability index (D) obtained through a frequency analysis of the longitude of the outer planet over two consecutive time intervals of 1000 yr. Red areas correspond to high orbital diffusion (instability) and the blue ones to low diffusion (stable orbits). The labeled contour plot shows the value of $\chi ^2$ obtained for each choice of parameters; the line without label indicates $\chi ^2 = 450$ . There are two different types of blue zones where the system can be stabilized, one of them corresponding to the 2:1 mean motion resonance (for a_d > 0.915 AU and e_d > 0.15).
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.4cm,clip]{6194fig8.ps}\par \end{figure}$ Figure 8: Orbital evolution of the four planets with a 1:1 resonance: inclination is equal to $90^\circ$ and a_b = 1.47 AU, e_b = 0.128, $\lambda _{0b} = 36.59^\circ$ , $\omega _b = 255.68^\circ$ , K_b = 62.00 m/s; a_c = 0.09 AU, e_c = 0.175, $\lambda _{0c} = 176.60^\circ$ , $\omega _c = 206.04^\circ$ , K_c = 3.20 m/s; a_d = 1.42 AU, e_d = 0.122, $\lambda _{0d} = 235.13^\circ$ , $\omega _d = 358.80^\circ$ , K_d = 27.94 m/s; and a_(e) = 3.93 AU, e_(e) = 0.350, $\lambda _{0{(e)}} = 89.79^\circ$ , $\omega _{(e)} = 164.00^\circ$ , K_(e) = 18.81 m/s. The panel shows a face-on view of the system; x and y are spatial coordinates in a frame centered on the star. Due to the strong interactions, the orbits of the Trojan planets undergo large variations and the system is destroyed in only 92 years. A dynamical analysis of the stability around the Trojan planets also provides no alternative stable possibility.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=8.5cm,clip]{6194fig1.eps}\par \end{figure}$	Figure 1: Phase-folded radial-velocity measurements (in km s^-1) and best fit for the two planets in the two-planet solution. For every planet the contribution by the other planets has been subtracted. In the upper panel showing the long-period planet, a time-correlated structure becomes evident. HARPS data are plotted in red, UCLES data in blue, and CORALIE data in green.
Open with DEXTER

$\begin{figure} \includegraphics[angle=270,width=8.8cm,clip]{6194fig3.ps}\par \end{figure}$	Figure 3: Global solution containing 4 planets and covering the full time range of observations. The lower plot shows the residuals of the fitted function to the actual measurements. The fit takes planet-planet interaction into account. The weighted dispersion of the residuals is 1.75 m s^-1.
Open with DEXTER

$\begin{figure} \includegraphics[width=8.3cm,clip]{6194fig4.eps}\par \end{figure}$	Figure 4: The figure shows a zoom on the time span covered by the HARPS data. It must be note that HARPS data constrain in an exceptional way the three shorter-period orbit. It particular, these complex curve would be incompatible even with a two-planet solution.
Open with DEXTER

$\begin{figure} \includegraphics[width=8cm,clip]{6194fig5.eps}\par \end{figure}$	Figure 5: The upper panel shows the phase-folded HARPS data and the best Keplerian best fit to the P=9.64 days-planet after subtracting the contribution of the other planets. The lower panels shows the residuals of the data to the best fit. The dispersion of the HARPS residuals is 1.41 m s^-1 rms.
Open with DEXTER