All Figures

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f1.eps}\end{figure}$ Figure 1: Average IUE spectrum of the Seyfert 1 galaxy Mrk 335. The position of the 12 spectral windows is indicated by the gray areas.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=16.7cm,clip]{1903f2.eps}\end{figure}$ Figure 2: Light curves in the range 1300-1350 Å for the 15 objects of our sample. Only one error bar is drawn on each panel for clarity (in the upper- or lower-left corner).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=17cm,clip]{1903f3.eps}\end{figure}$ Figure 3: Structure functions of the light curves at 1300-1350 Å. The horizontal dashed lines show the asymptotic value of twice the variance of x(t). The continuous line shows the best fit, while the value of $\tau _{\rm max}$ is shown with its uncertainty above the SF to ease readability. The vertical dot-long dash lines indicate the location of the characteristic time scales found in CP01, while the short dash-long dash line indicates the value found in Paltani et al. (1998).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f4.eps}\end{figure}$ Figure 4: Histogram of the event durations $2\mu _{\rm sim, 1300}$ input to the simulation that produce a measured $\tau _{\rm max}$ in the range 0.4-0.5 years, for 3C 273. The inset shows the same diagram in linear scale, fitted with a Gaussian.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=17cm,clip]{1903f5.eps}\end{figure}$ Figure 5: SFs for the light curves 1300 to 3000 Å, for NGC 5548. The horizontal dashed lines show the asymptotic value of twice the variance of the light curve. The best fits are shown by a continuous line.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=14.2cm,clip]{1903f6.eps}\end{figure}$ Figure 6: $\tau _{\rm max}$ as a function of wavelength for the 15 objects. The $\tau _{\rm max}$ are corrected for time dilatation. The two objects which present a very strong increase in $\tau _{\rm max}$ at long wavelengths are shown with a dashed line.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f7.eps}\end{figure}$ Figure 7: Variability time scales $2\mu _{1300}$ as a function of the luminosity of the objects in the range 1300-1350 Å. The BCES regression is shown with the dashed line.
Open with DEXTER

In the text

$\begin{figure}\par\includegraphics[width=8.6cm,clip]{1903f8.eps}\end{figure}$ Figure 8: Event amplitude spectrum $H_{\lambda }$ as a function of wavelength for NGC 5548, with $C_{\lambda }=0$ (squares), and $C_{\lambda }=L_\lambda ^{{\rm min}}$ (open triangles). The lower curve has been slightly shifted to the right to ease readability.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[bb=20 150 570 500,clip,width=8.8cm,clip]{1903f9.eps}\end{figure}$ Figure 9: Event energy as a function of the object luminosity in the hypothesis $C_{\lambda }=0$ ( left panel), and using the upper limit of $C_{\lambda }$ ( right panel). The BCES regression is shown with the dashed line.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[bb=20 150 570 500,clip,width=8.8cm,clip]{1903f10.eps}\end{figure}$ Figure 10: Event rate as a function of the object luminosity in the hypothesis $C_{\lambda }=0$ ( left panel), and using the upper limit of $C_{\lambda }$ ( right panel).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f11.eps}\end{figure}$ Figure 11: Time scales vs. black hole mass. The gray area shows the range of timescales found in this study.
Open with DEXTER

In the text

$\begin{figure}\par\includegraphics[width=15cm,clip]{m335.4.4.eps}\end{figure}$ Figure A.1: Mrk 335, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the SF (d), (f), (j), (l), because of noisy light curves (see Figs. A.2-A.5). The time scales corresponding to the SF (d), (f), (j), (l) are thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m335.lc.1.eps}\end{figure}$ Figure A.2: Mrk 335, comparison of the noisy light curve 1950-2000 Å (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m335.lc.2.eps}\end{figure}$ Figure A.3: Mrk 335, comparison of the noisy light curve 2200-2250 Å (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m335.lc.3.eps}\end{figure}$ Figure A.4: Mrk 335, comparison of the noisy light curve 2700-2750 Å (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m335.lc.4.eps}\end{figure}$ Figure A.5: Mrk 335, comparison of the noisy light curve 2975-3025 Å (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{m509.4.4.eps}\end{figure}$ Figure A.6: Mrk 509, SF for the light curves 1300 to 3000 Å.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{m926.4.4.eps}\end{figure}$ Figure A.7: Mrk 926, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the all the SF built from data taken with the LW instrument (1950-3025 Å). The reason is the small number of epochs (16) taken with the LW (see Fig. A.8). We thus reject the time scales corresponding to the SF (d)-(l).

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m926.lc.1.eps}\end{figure}$ Figure A.8: Mrk 926, comparison of the LW light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{m1095.4.4.eps}\end{figure}$ Figure A.9: Mrk 1095, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the SF (d), (e), (f), because of noisy light curves (see Fig. A.10). The time scales corresponding to the SF (d), (e), (f) are thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{m1095.lc.1.eps}\end{figure}$ Figure A.10: Mrk 1095, comparison of the noisy 1950-2000, 2100-2150, 2200-2250 light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{n3516.4.4.eps}\end{figure}$ Figure A.11: NGC 3516, SF for the light curves 1300 to 3000 Å. The fit of the SF (a)-(c) follows what was done in Sect. 4.1 (two parameters fixed). The LW was less used than the SWP during the intensive monitoring made in 1993 (see Fig. A.12). The number of LW epochs is smaller (22) than for the SWP (71). Although the fit seems to have succeeded in the LW range, one should be careful in interpreting the result as evidence for the presence of a longer timescale as the SF slope $\alpha$ pegged to the maximum value of 2. In addition, the fit is rather poor. We thus decided to reject the time scales corresponding to the SF (d)-(l).

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{n3516.lc.1.eps}\end{figure}$ Figure A.12: NGC 3516, comparison of the LW light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{n3783.4.4.eps}\end{figure}$ Figure A.13: NGC 3783, SF for the light curves 1300 to 3000 Å. The time scales obtained from the LW data are more scattered because the LW data have 12% less epochs than the SWP.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{n4151.4.4.eps}\end{figure}$ Figure A.14: NGC 4151, SF for the light curves 1300 to 3000 Å. The fit of the SF follows what was done in Sect. 4.1 (two parameters fixed). The fit did not succeed for the SF (d), the corresponding time scale was thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{n4593.4.4.eps}\end{figure}$ Figure A.15: NGC 4593, SF for the light curves 1300 to 3000 Å.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{5548.0.9.eps}\end{figure}$ Figure A.16: NGC 5548, SF for the light curves 1300 to 3000 Å.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{n7469.4.4.eps}\end{figure}$ Figure A.17: NGC 7469, SF for the light curves 1300 to 3000 Å. The fit of the SF (a)-(c) follows what was done in Sect. 4.1 (two parameters fixed). Only SWP was used for the intensive monitoring made in 1996 (see Fig. A.18). As in NGC 3516, the number of LW epochs is smaller (16) than for the SWP (95). The fit did not succeed for the SF (e), (f), (h), (k), (l). The corresponding time scales were thus rejected. The SF (d), (g), (i), (j) may be interpreted as showing evidence for the presence of a longer timescale in the data, but this result should be taken with caution because of the small number of epochs taken into account.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{n7469.lc.1.eps}\end{figure}$ Figure A.18: NGC 7469, comparison of the LW light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{3c120.4.4.eps}\end{figure}$ Figure A.19: 3C 120.0, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the SF (b), (d), (g)-(l) because of noisy light curves (see Fig. A.20, A.21). The time scales corresponding to these SF are thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c120.lc.1.eps}\end{figure}$ Figure A.20: 3C 120.0, comparison of the noisy 1450-1500 Å light curve (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c120.lc.2.eps}\end{figure}$ Figure A.21: 3C 120.0, comparison of the noisy LW (d), (g)-(l) light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{3c273.4.4.eps}\end{figure}$ Figure A.22: 3C 273, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the SF (i)-(k) because of noisy light curves (see Fig. A.23-A.25). The time scales corresponding to these SF are thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c273.lc.1.eps}\end{figure}$ Figure A.23: 3C 273, comparison of the noisy 2550-2600 Å light curve (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c273.lc.2.eps}\end{figure}$ Figure A.24: 3C 273, comparison of the noisy 2700-2750 Å light curve (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c273.lc.3.eps}\end{figure}$ Figure A.25: 3C 273, comparison of the noisy 2875-2925 Å light curve (dashed line) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{3c390.4.4.eps}\end{figure}$ Figure A.26: 3C 390.3, SF for the light curves 1300 to 3000 Å. Only SWP was used for the intensive monitoring made in 1995-1996 (see Fig. A.27). The number of LW epochs is too small to build measurable SF. The fit did not succeed and no time scales could be deduced in the range 1950 to 2975 Å for this object.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{3c390.lc.1.eps}\end{figure}$ Figure A.27: 3C 390.3, comparison of the LW light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{f9.4.4.eps}\end{figure}$ Figure A.28: Fairall 9, SF for the light curves 1300 to 3000 Å. The fit of the SF (a)-(c) follows what was done in Sect. 4.1 (two parameters fixed). The number of epochs between SWP and LW is very uneven (139 for the SWP and 63 for the LW). The SF (d)-(l) may be interpreted as showing evidence for the presence of a longer timescale in the data.

In the text

$\begin{figure} \par\includegraphics[width=15cm]{eso141.4.4.eps}\end{figure}$ Figure A.29: ESO 141-55, SF for the light curves 1300 to 3000 Å. The fit did not succeed for the SF (d), (f) due to noisy light curves (see Fig. A.30); the corresponding time scales were thus rejected.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm]{eso141.lc.1.eps}\end{figure}$ Figure A.30: ESO 141-55, comparison of the noisy 1950-2000 Å and 2200-2250 Å light curves (dashed lines) with the 1300-1350 Å light curve.

In the text

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f1.eps}\end{figure}$	Figure 1: Average IUE spectrum of the Seyfert 1 galaxy Mrk 335. The position of the 12 spectral windows is indicated by the gray areas.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=16.7cm,clip]{1903f2.eps}\end{figure}$	Figure 2: Light curves in the range 1300-1350 Å for the 15 objects of our sample. Only one error bar is drawn on each panel for clarity (in the upper- or lower-left corner).
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f4.eps}\end{figure}$	Figure 4: Histogram of the event durations $2\mu _{\rm sim, 1300}$ input to the simulation that produce a measured $\tau _{\rm max}$ in the range 0.4-0.5 years, for 3C 273. The inset shows the same diagram in linear scale, fitted with a Gaussian.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=17cm,clip]{1903f5.eps}\end{figure}$	Figure 5: SFs for the light curves 1300 to 3000 Å, for NGC 5548. The horizontal dashed lines show the asymptotic value of twice the variance of the light curve. The best fits are shown by a continuous line.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=14.2cm,clip]{1903f6.eps}\end{figure}$	Figure 6: $\tau _{\rm max}$ as a function of wavelength for the 15 objects. The $\tau _{\rm max}$ are corrected for time dilatation. The two objects which present a very strong increase in $\tau _{\rm max}$ at long wavelengths are shown with a dashed line.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f7.eps}\end{figure}$	Figure 7: Variability time scales $2\mu _{1300}$ as a function of the luminosity of the objects in the range 1300-1350 Å. The BCES regression is shown with the dashed line.
Open with DEXTER

$\begin{figure}\par\includegraphics[width=8.6cm,clip]{1903f8.eps}\end{figure}$	Figure 8: Event amplitude spectrum $H_{\lambda }$ as a function of wavelength for NGC 5548, with $C_{\lambda }=0$ (squares), and $C_{\lambda }=L_\lambda ^{{\rm min}}$ (open triangles). The lower curve has been slightly shifted to the right to ease readability.
Open with DEXTER

$\begin{figure} \par\includegraphics[bb=20 150 570 500,clip,width=8.8cm,clip]{1903f9.eps}\end{figure}$	Figure 9: Event energy as a function of the object luminosity in the hypothesis $C_{\lambda }=0$ ( left panel), and using the upper limit of $C_{\lambda }$ ( right panel). The BCES regression is shown with the dashed line.
Open with DEXTER

$\begin{figure} \par\includegraphics[bb=20 150 570 500,clip,width=8.8cm,clip]{1903f10.eps}\end{figure}$	Figure 10: Event rate as a function of the object luminosity in the hypothesis $C_{\lambda }=0$ ( left panel), and using the upper limit of $C_{\lambda }$ ( right panel).
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.8cm,clip]{1903f11.eps}\end{figure}$	Figure 11: Time scales vs. black hole mass. The gray area shows the range of timescales found in this study.
Open with DEXTER