Calibration Report for the Side-Looking Imager Strips
          of the Descent Imager/Spectral Radiometer
 Instrument aboard the Huygens Probe of the Cassini Mission
                              
                              
                             by
                         L. R. Doose
                              
                        22 June 2006

1.0  Introduction

Titan's atmosphere may contain "layers", or at least
concentrations of aerosols at favored altitudes.  This
layering might be caused by the condensation of different
species or aerosols or by the vertical transport process of
the aerosols.  Changes in the nadir angle dependence of the
intensity in the DISR side-looking imager (SLI) are a
sensitive way to detect such layers.

SLI images are exposed relatively infrequently during the
descent, however.  It is possible to fall several kilometers
between them.  In order to increase the altitude resolution
of such measurements a special type of processing of SLI
data was devised.  The resulting data are called SLI strips.

These data also can provide useful constraints on the
attitude of the probe.  The shape of the intensity vs. nadir
angle is helpful in determining how much the probe is
tipped.



2.0  Pixel Layout

The SLI strips data are taken from pixels near both vertical
edges of the SLI.  The SLI includes 128 CCD columns.  The
first strip of pixels is taken from columns 6 through 18,
and the second set are from columns 109 through 123.  The
extreme edges of the SLI are avoided, because large numbers
of defective pixels are concentrated there.



3.0  Flight Software Processing

The 13 columns on each side of the SLI are summed for each
row.  The strip data therefore consist of an array of 2
columns x 254 rows of data.  Unfortunately due to an error
in the flight software, the data are shifted in rows.  In
the first column the data values for the first two rows are
erroneously set to zero.  In this column the true first row
of data is offset to row 3, the second row to row 4, etc.
The last two rows of data are lost.  The situation is
similar for the second column, except the offset is one row
instead of two.  The first row is erroneously set to zero,
and the last row of data is lost.

Unlike normal imager data there is no:

Subtraction of 8 data numbers (DN)
Flat field correction
Square-root transformation
DCS bad pixel replacement
DCS image compression

Because the data are not converted to 8-bits/pixel by square-
root processing and not compressed, it is inherently more
accurate than SLI image data.



4.0  Computation of Radiance from Data Number

The mean radiance over the 13 summed pixels in each row may
be computed using:

1.   Compute the dark current in each of the pixels from the
      dark current model (see Dark Current Estimation for the CCD
      of the Descent Imager/Spectral Radiometer aboard the Huygens
      Probe of the Cassini Mission)
2.   Subtract the dark current  from the 13 summed pixels in
      each row
3.   Divide by the exposure time in seconds
4.   Divide by the absolute responsivity of each imager
      pixel at the correct CCD temperature
5.   Divide by 13

The absolute responsivities can be found in "Calibration
Report for the Imagers of the Descent Imager/Spectral Radiometer 
Instrument aboard the Huygens Probe of the Cassini Mission."  

This will yield the mean radiance in the SLI imager bandpass.  The
wavelengths of this bandpass can be found in the same
document.