All Figures

$\begin{figure} \par\psfig{figure=2767fg1.ps,height=7.2cm,width=8.5cm,clip=}\end{figure}$ Figure 1: Background subtracted, 0.6-10 keV PN light curves of Mrk 421 with time binning of 100 s. Times are counted from the beginning of the actual exposure. The curves are labeled by the orbit of the observation.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg2a.ps,height=6.5truecm,width=8.5truecm,an... ...cm,angle=-90,% bbllx=383pt,bblly=40pt,bburx=560pt,bbury=709pt,clip=}\end{figure}$ Figure 2: Broken power law model fit to the 0.6-10 keV data during the three observations. The two top figures show the fit and the data to model ratio for orbit 84, the middle panel the ratio for orbit 546, the bottom panel that for orbit 807.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg3.eps,height=9.5truecm,width=8.5truecm,angle=0,% bbllx=34pt,bblly=254pt,bburx=508pt,bbury=756pt,clip=}\end{figure}$ Figure 3: CCF plots for the orbit 84 ( top), orbit 546 ( middle) and orbit 807 ( bottom) observation. The 10 s binned soft and hard band light curves for the total observations were used; the error bars are smallerthan the symbol sizes. The curves are labeled by the orbit of theobservation.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg4a.ps,height=4.4truecm,width=8.5truecm,an... ...5truecm,angle=0,% bbllx=11pt,bblly=1pt,bburx=485pt,bbury=85pt,clip=}\end{figure}$ Figure 4: Sliding window CCFs for orbit 546. The amplitude of the CCF is color coded, with lags plotted in the vertical direction. The top figure was calculated with an individual data length of 2000 s, the middle with 6000 s, and the bottom figure with 20 000 s long data streams. The bottom panel shows the color bar used for the numerical values of the CCFs for the above figure and for Fig. 6.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg5.ps,height=4.5truecm,width=8.6truecm,angle=0,% bbllx=6pt,bblly=6pt,bburx=415pt,bbury=196pt,clip=}\end{figure}$ Figure 5: Standard deviations $\sigma _{P({\rm ccf})}$ of the sliding window CCFs for orbit 546 for different data lengths $\mathcal{L}$ . The diamonds are the measured values, the three large symbols indicate the window lengths used for the CCFs in Fig. 4. The solid line gives the mean and the dashed lines the standard deviation of the re-sampled CCFs (see text).
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg6a.ps,height=3.48truecm,width=8.5truecm,a... ...ruecm,angle=0,% bbllx=32pt,bblly=28pt,bburx=436pt,bbury=220pt,clip=}\end{figure}$ Figure 6: Sliding window CCFs for orbits 084 ( top), orbit 546 ( middle) and orbit 807 ( bottom) calculated with the optimized sliding window lengths (see text). The amplitude of the CCF is color coded, with lags plotted in the vertical direction. Plotted below the individual CCFs are the corresponding light curves in the total (0.6-10 keV) energy band. The x-axis is the time from the beginning of the observation which, for the CCFs, corresponds to the start time of the data window for the calculations.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg7.eps,height=8.5cm,width=8.5cm,clip=}\end{figure}$ Figure 7: Background subtracted, 0.6-2 and 4-10 keV PN light curves of Mrk 421 in orbit 84 (open circles and filled squares, respectively), in bins of size 200 s ( top panel). The light curves are normalized to their mean. The corresponding $\Gamma _{\rm hard}$ and $\Gamma _{\rm soft}$ curves (filled squares and open circles, respectively) are plotted in the bottom panel (binned as described in the text). The shaded boxes identify those parts of the observation where the $\Gamma$ vs. source flux relation is studied in more detail (see Sect. 4.1).
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg8.eps,height=8.5cm,width=8.5cm,clip=}\end{figure}$ Figure 8: Same as Fig. 7, but for the observation during orbit 546. The hard band and soft band count rate light curves are in bins of size 1000 s and 500 s, respectively.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg9.eps,height=8.5cm,width=8.5cm,clip=}\end{figure}$ Figure 9: Same as Fig. 7, but for the observation during orbit 807. The count rate light curves are in bins of size 40 s.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{2767fg10.eps}\end{figure}$ Figure 10: The $\Gamma _{\rm hard}$ vs. 4-10 keV count rate and the $\Gamma _{\rm soft}$ vs. 0.6-2 keV relations ( top and middle panels, respectively). The bottom panel shows the ratio $\Gamma _{\rm hard}/\Gamma _{\rm soft}$ plotted as a function of the 0.6-10 keV count rate. Data points from Figs. 7-9 are shown with open circles, open squares and filled triangles, respectively. The solid curves describe broad trends as discussed in the text. For the orbit 84 observation, which was performed with a thick filter, the 0.6-2, 4-10 and 0.6-10 keV count rates are multiplied by factors of 1.22, 1.01, and 1.19, respectively. These are the count rate ratios expected for the thick and medium thick filters in the case of a power law spectrum and $N_{\rm H}$ similar to that of Mrk 421.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{2767fg11.eps}\end{figure}$ Figure 11: The average cross-correlation function between $\Gamma _{\rm hard}$ and 4-10 keV count rate ( top panel), the $\Gamma _{\rm soft}$ and 0.6-2 keV count rate ( middle panel), and between the two spectral slopes $\Gamma _{\rm hard}$ and $\Gamma _{\rm soft}$ ( bottom panel), using data from all observations. The CCFs are defined such that any positive delays would imply a delay of the spectral with respect to the flux variations (in the upper two panels), and a delay of the $\Gamma _{\rm hard}$ with respect to the $\Gamma _{\rm soft}$ variations, in the bottom panel.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg12.eps,height=9.5truecm,width=8.5truecm,clip=}\end{figure}$ Figure 12: Orbit 84: the $\Gamma _{\rm hard}$ vs. 4-10 keV count rate and the $\Gamma _{\rm soft}$ vs. 0.6-2 keV relations ( top and middle panels, respectively) for the "P1'' and "P2'' parts (identified by the shaded boxes in Fig. 7). The bottom panel shows the ratio $\Gamma _{\rm hard}/\Gamma _{\rm soft}$ as a function of the 0.6-10 keV count rates. Filled circles show the data points during the flux rise and open circles during the decaying phase of each event. The dashed line indicates the linear path that the rising (or the decaying) phase points follow. For clarity, in cases where the two phase points follow different paths, or define loop-like structures, the respective points are connected with solid lines.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg13.eps,height=9.5truecm,width=8.5truecm,clip=}\end{figure}$ Figure 13: Same as Fig. 12, but for the parts "P1'' and "P2" in orbit 546.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg14.eps,height=9.5truecm,width=8.5truecm,clip=}\end{figure}$ Figure 14: Same as Fig. 12, but for the parts "P1'' and "P2'' in orbit 807.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg15.eps,height=6.8truecm,width=8.5truecm,angle=0,% bbllx=3pt,bblly=31pt,bburx=708pt,bbury=530pt,clip=}\end{figure}$ Figure 15: Density (in arbitrary units) of the colliding shells at different evolution times (in the jet frame) of the system as seen by a distant observer. The observation time (with arbitrary origin) runs from right to left. The plot shows at what times the observer would "see'' the various structures approaching him with relativistic velocities.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg16.eps,height=6.2truecm,width=8.5truecm,angle=0,% bbllx=18pt,bblly=39pt,bburx=789pt,bbury=525pt,clip=}\end{figure}$ Figure 16: Normalized X-ray emission from the model of Fig. 15 in the top panel in the hard (1-10 keV, red curve) and the soft (0.1-1 keV) X-ray band. The bottom panel shows the evolution of the spectral power law indices in the two bands.
Open with DEXTER

In the text

$\begin{figure} \par\psfig{figure=2767fg1.ps,height=7.2cm,width=8.5cm,clip=}\end{figure}$	Figure 1: Background subtracted, 0.6-10 keV PN light curves of Mrk 421 with time binning of 100 s. Times are counted from the beginning of the actual exposure. The curves are labeled by the orbit of the observation.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg2a.ps,height=6.5truecm,width=8.5truecm,an... ...cm,angle=-90,% bbllx=383pt,bblly=40pt,bburx=560pt,bbury=709pt,clip=}\end{figure}$	Figure 2: Broken power law model fit to the 0.6-10 keV data during the three observations. The two top figures show the fit and the data to model ratio for orbit 84, the middle panel the ratio for orbit 546, the bottom panel that for orbit 807.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg3.eps,height=9.5truecm,width=8.5truecm,angle=0,% bbllx=34pt,bblly=254pt,bburx=508pt,bbury=756pt,clip=}\end{figure}$	Figure 3: CCF plots for the orbit 84 ( top), orbit 546 ( middle) and orbit 807 ( bottom) observation. The 10 s binned soft and hard band light curves for the total observations were used; the error bars are smallerthan the symbol sizes. The curves are labeled by the orbit of theobservation.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg5.ps,height=4.5truecm,width=8.6truecm,angle=0,% bbllx=6pt,bblly=6pt,bburx=415pt,bbury=196pt,clip=}\end{figure}$	Figure 5: Standard deviations $\sigma _{P({\rm ccf})}$ of the sliding window CCFs for orbit 546 for different data lengths $\mathcal{L}$ . The diamonds are the measured values, the three large symbols indicate the window lengths used for the CCFs in Fig. 4. The solid line gives the mean and the dashed lines the standard deviation of the re-sampled CCFs (see text).
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg8.eps,height=8.5cm,width=8.5cm,clip=}\end{figure}$	Figure 8: Same as Fig. 7, but for the observation during orbit 546. The hard band and soft band count rate light curves are in bins of size 1000 s and 500 s, respectively.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg9.eps,height=8.5cm,width=8.5cm,clip=}\end{figure}$	Figure 9: Same as Fig. 7, but for the observation during orbit 807. The count rate light curves are in bins of size 40 s.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg13.eps,height=9.5truecm,width=8.5truecm,clip=}\end{figure}$	Figure 13: Same as Fig. 12, but for the parts "P1'' and "P2" in orbit 546.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg14.eps,height=9.5truecm,width=8.5truecm,clip=}\end{figure}$	Figure 14: Same as Fig. 12, but for the parts "P1'' and "P2'' in orbit 807.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg15.eps,height=6.8truecm,width=8.5truecm,angle=0,% bbllx=3pt,bblly=31pt,bburx=708pt,bbury=530pt,clip=}\end{figure}$	Figure 15: Density (in arbitrary units) of the colliding shells at different evolution times (in the jet frame) of the system as seen by a distant observer. The observation time (with arbitrary origin) runs from right to left. The plot shows at what times the observer would "see'' the various structures approaching him with relativistic velocities.
Open with DEXTER

$\begin{figure} \par\psfig{figure=2767fg16.eps,height=6.2truecm,width=8.5truecm,angle=0,% bbllx=18pt,bblly=39pt,bburx=789pt,bbury=525pt,clip=}\end{figure}$	Figure 16: Normalized X-ray emission from the model of Fig. 15 in the top panel in the hard (1-10 keV, red curve) and the soft (0.1-1 keV) X-ray band. The bottom panel shows the evolution of the spectral power law indices in the two bands.
Open with DEXTER