Processed data cubes of observations from October 16, 2004
  Target was AR 0682, located at S14E31 (mu=0.8)

General:
   All cubes are aligned with respect to the lower left corner.  The
   size of the cubes has not been cropped to identical dimensions,
   because else the very interesting pore at the upper border would
   hardly be in the FOV due to a relatively high shift of the Ca
   images.
   All images have been binned 2x2, the pixel size is 0.142 arcsec
   square, the common (=smallest) area is 77.2x55.7 arcsec.
   The cubes are stored as integer numbers, in arbirtray units.
   The observations span the time between 8:34 and 9:09 UT, the
   cadence was 45s.  The series contains two gaps, the first between
   frame 8 and 9, the other between 30 and 31 (counting starts with
   0).  There, a frame is missing, thus the time difference between
   the frames is 90s.
   There are no separate cubes for red continuum and H-alpha, as they
   are combined in a 4D cube 544x412x6x46.  the 3rd dimension holds
   the different wavelength images: the first is the red continuum
   image, then come the H-alpha images, starting with the blue wing
   image and advancing towards longer wavelengths.

Data processing:

   For all broad-band cubes, data processing included the speckle
   reconstruction using the Speckle Masking method.  The single frames
   are aligned using cross correlation techniques in fourier space to
   measure the shift.
   Remaining seeing influence due to incomplete reconstruction is compensated
   by equalizing the radially averaged powerspectra.
   G Band serves as reference for the rotation and scaling alignment, as the
   different channels have not exactly the same orientation and magnification
   (<0.5 deg rotation ans < 0.2% scale).  rotation/scaling is done via IDL
   routine ROT with keyword CUBIC=-0.5
   The red continuum reconstruted image is used to process the
   narrow-band H-alpha images using the Keller&v.d.Luehe method.

Orientation:
  For all cubes north is up, east is left.  This is sky north, not solar
  north.  P0/B0 for that day were 26.1 / 5.8 degree

Ca data:
  Taken with narrow band interference filter centered on Ca II H line core
  (396.849 nm).  Filter width is 0.135 nm.

G Band:
  Taken with broad band (1 nm) interference filter centered on the G Band
  spectral feature (430.5 nm)

Red Continuum
  Taken with broad band (0.5 nm) interference filter centered in a
  quite clean continuum area at 655 nm.

H-alpha:
  Taken with a narrow-band (0.025nm) tunable Lyot filter.  The filter
  was scanning the line profile at 5 wavelength positions: 
  -0.070, -0.035, 0, +0.035 and +0.070 nm relative to the line core

Velocity:
  computed from blue and red wing images using V=F((b-r)/(b+r)) where
  the calibration function F is the theoretical response function,
  computed using an atlas profile and the (known) spectral profile of
  the Lyot filter.  The values of this map are in m/s.