All Figures

$\begin{figure} \par\includegraphics[width=8.4cm,clip]{7330fi01.eps} \end{figure}$ Figure 1: Continuum intensity image of NOAA 10436 at $15~666.5\pm0.5$ Å. White (black) contours mark the umbral/penumbral (penumbral/quiet Sun) boundaries. The straight black line marks the slit position during the time series and the white portion indicates the region studied in detail. The dark grey pixel within the white portion of the slit marks the position from which the profiles in Fig. 2 are taken, while the arrow in the umbra points to disk centre.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=9cm,clip]{7330fi02.eps} \end{figure}$ Figure 2: Example penumbral profiles of Stokes I a), Stokes Q b), Stokes U c), and Stokes V d), each normalized to the local continnum intensity, $I_{\rm C}$ , from the pixel marked dark grey inside the region of interest shown in Fig. 1. Wavelengths are relative to the 15 662 Å line, with dotted lines at the blueshift-corrected laboratory values.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi03.eps} \end{figure}$ Figure 3: Space-time plots of continuum intensity as a percentage of average quiet-Sun continnum a), magnitude of the relative Stokes V area asymmetry b), and Stokes Q at +0.145 Å from the core of the 15 662 Å line c). d) Spatially-averaged Stokes Q signal from the region of interest. Only the slit portion extending toward solar north east from the northern umbra is shown in a)- c). The white (black) contour marks the umbral/penumbral (penumbral/quiet Sun) boundary and the dot-dashed lines bound the region studied (white area of slit in Fig. 1).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi04.eps} \end{figure}$ Figure 4: Variation through the limb-side penumbra of the parameters obtained by the inversion at $\log \left( \tau \right) = 0$ : temperature a), field strength b), local solar inclination c), and filling factor d). Points mark the temporal mean for each spatial pixel and error bars extend over $\pm$ 1 $\sigma$ .
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi05.eps} \end{figure}$ Figure 5: Height variation of wavelength-integrated 15 662 Å and 15 665 Å Stokes Q velocity response a), total pressure b), acoustic c), and Alfvénic d) wave speeds for the magnetic background and flux tube atmospheres (solid and dotted curves, respectively). Vertical lines in a) mark the COG in each component; vertical dashed and dot-dashed lines in b)- d) show these heights translated into the reference frame of the other atmosphere by enforcing total pressure balance.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=12cm,clip]{7330fi06.eps} \end{figure}$ Figure 6: a) Fourier power spectra from co-spatial magnetic background and flux tube atmosphere LOS velocities (solid and dotted curves, respectively). b) Fourier phase difference spectra between the magnetic background and flux tube velocities from the eleven analyzed pixels. Darker shading denotes greater Fouriercoherence and larger symbol size greater cross-spectral power. c) PDF of phase difference values over the range 2.5-4.5 mHz. The thick curve displays the measured values and the thin curve the best-fit Gaussian profile to the data.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=7.4cm,clip]{7330fi07.eps} \end{figure}$ Figure 7: Variation of propagation angle from the vertical, $\psi$ , with angular wave mode, l, for acoustic waves at a frequency of 3.5 mHz. The solid curve shows the case 50 km below the equipartition level (where the acoustic and Alfvén speeds are equal) in a 1.75 kG field inclined at 20 $^\circ$ , while the dashed curve is 90 km below the equipartition level in a 1.5 kG field inclined at 75 $^\circ$ (the sampling heights and configurations of the magnetic background and flux tube components, respectively).
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi08.eps} \end{figure}$ Figure 8: Comparison of RF-predicted (solid lines) and calculated (dotted lines) wave travel distances in the reference frame of the magnetic background (left) and the flux tube (right). Cases are presented for field-aligned waves propagating at the Alfvén (top) and slow-mode tube speeds (bottom). The RF-predicted vertical height separations have been converted into path length along the propagation direction.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi09.eps} \end{figure}$ Figure 9: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating vertically.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi10.eps} \end{figure}$ Figure 10: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 40 $^\circ$ to the vertical.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi11.eps} \end{figure}$ Figure 11: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 50 $^\circ$ to the vertical.
Open with DEXTER

In the text

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi12.eps} \end{figure}$ Figure 12: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 60 $^\circ$ to the vertical.
Open with DEXTER

In the text

$\begin{figure} \includegraphics[width=9cm,clip]{7330fi02.eps} \end{figure}$	Figure 2: Example penumbral profiles of Stokes I a), Stokes Q b), Stokes U c), and Stokes V d), each normalized to the local continnum intensity, $I_{\rm C}$ , from the pixel marked dark grey inside the region of interest shown in Fig. 1. Wavelengths are relative to the 15 662 Å line, with dotted lines at the blueshift-corrected laboratory values.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi04.eps} \end{figure}$	Figure 4: Variation through the limb-side penumbra of the parameters obtained by the inversion at $\log \left( \tau \right) = 0$ : temperature a), field strength b), local solar inclination c), and filling factor d). Points mark the temporal mean for each spatial pixel and error bars extend over $\pm$ 1 $\sigma$ .
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi09.eps} \end{figure}$	Figure 9: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating vertically.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi10.eps} \end{figure}$	Figure 10: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 40 $^\circ$ to the vertical.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi11.eps} \end{figure}$	Figure 11: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 50 $^\circ$ to the vertical.
Open with DEXTER

$\begin{figure} \par\includegraphics[width=8.5cm,clip]{7330fi12.eps} \end{figure}$	Figure 12: As for Fig. 8, but for acoustic (top) and fast-mode waves (bottom) propagating at 60 $^\circ$ to the vertical.
Open with DEXTER