$\begin{figure} \par\includegraphics[angle=-90,width=8.8cm]{8794fig1} \end{figure}$ Figure 1: Fractional errors on the total energy E with respect to the initial one E₀ (black curve), and on the total angular momentum H with respect to the initial one H₀ (grey curve), as a function of time over the 10⁸ yr integration.
Open with DEXTER

In the text

$\begin{figure} \par\makebox[18cm]{ \includegraphics[angle=-90,width=6cm]{8794fg2... ...]{8794fg2e}\hfil \includegraphics[angle=-90,width=6cm]{8794fg2f} }\end{figure}$ Figure 2: The 10⁴ first years of the integration of the nominal solution. The upper plots show the temporal evolution of the eccentricity of the three planets Gl 581 b), c), and d), from left to right, respectively; the lower plots are the same for the longitude of periastron $\varpi$ .
Open with DEXTER

In the text

$\begin{figure} \par\makebox[17.5cm]{ \includegraphics[angle=-90,width=8.2cm]{879... ...pace*{5mm} \includegraphics[angle=-90,width=8.2cm]{8794fg3d} }\par\end{figure}$ Figure 3: Stability of the three-planet system in various configurations. The maximum variation range for the semi-major axes ( upper plots) and for the eccentricities ( lower plots) is displayed as a function of the assumed viewing inclination of the system with respect to pole-on. Each cross corresponds to a single simulation. The left plots correspond to simulations with the nominal eccentricities as initial conditions, and the right plots to simulations with eccentricities increased by $1\sigma$ relative to the error bars given in Table 1.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[angle=-90,width=8.8cm]{8794fig4} \end{figure}$ Figure 4: Progress of the calculation of the partial Lyapunov exponents as a function of the integration time, in the nominal case for Gl 581 (zero inclination, nominal eccentricities). At t=10⁴ yr, the three exponents have stabilised.
Open with DEXTER

In the text

$\begin{figure} \par\makebox[18cm]{ \includegraphics[angle=-90,width=8.8cm]{8794fg5a}\hspace*{2mm} \includegraphics[angle=-90,width=8.8cm]{8794fg5b} }\end{figure}$ Figure 5: Lyapunov exponents as a function of the inclination i, computed for all the simulations described in Fig. 3. Left plot: simulations with nominal eccentricities; Right plot: simulations with $1\sigma$ increased eccentricities.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[angle=-90,width=8.8cm]{8794fig1} \end{figure}$	Figure 1: Fractional errors on the total energy E with respect to the initial one E₀ (black curve), and on the total angular momentum H with respect to the initial one H₀ (grey curve), as a function of time over the 10⁸ yr integration.
Open with DEXTER

$\begin{figure} \par\makebox[18cm]{ \includegraphics[angle=-90,width=6cm]{8794fg2... ...]{8794fg2e}\hfil \includegraphics[angle=-90,width=6cm]{8794fg2f} }\end{figure}$	Figure 2: The 10⁴ first years of the integration of the nominal solution. The upper plots show the temporal evolution of the eccentricity of the three planets Gl 581 b), c), and d), from left to right, respectively; the lower plots are the same for the longitude of periastron $\varpi$ .
Open with DEXTER

$\begin{figure} \par\includegraphics[angle=-90,width=8.8cm]{8794fig4} \end{figure}$	Figure 4: Progress of the calculation of the partial Lyapunov exponents as a function of the integration time, in the nominal case for Gl 581 (zero inclination, nominal eccentricities). At t=10⁴ yr, the three exponents have stabilised.
Open with DEXTER

$\begin{figure} \par\makebox[18cm]{ \includegraphics[angle=-90,width=8.8cm]{8794fg5a}\hspace*{2mm} \includegraphics[angle=-90,width=8.8cm]{8794fg5b} }\end{figure}$	Figure 5: Lyapunov exponents as a function of the inclination i, computed for all the simulations described in Fig. 3. Left plot: simulations with nominal eccentricities; Right plot: simulations with $1\sigma$ increased eccentricities.
Open with DEXTER