Summary of the VGISS peer review December 19, 2006.


Participants
------------

Steven Adams, Charlie Avis, Bonnie Buratti, Nancy Chanover, Mike Evans,
Joel Mosher, Kathy Rages, Mark Showalter, Mitch Gordon.


Overall peer review outcome
---------------------------

The status is, "The material did not pass peer review and may need to
undergo an incremental review in certain specified areas."

A critical issue was identified involving dark current subtraction (see
Issue 1. below). The Rings Node will attempt to identify the source of
the problem and propose an approach to rectify the problem. The results
of this effort will be distributed to the panel at which time the panel
will determine whether an incremental review of a portion of the data
set is required. If the panel decides that an incremental review is not
required, the status of the review will be upgraded to, "The material
passed peer review and can be archived once a set of panel identified
liens have been resolved."

Summary of key issues and associated liens
------------------------------------------

TERMINOLOGY: "Image Set". This archive contains multiple image files
reflecting different stages in image processing for each image. For this
document, we use the term "image set" to refer to all of the versions
(e.g., RAW, CLEANED, CALIB) of a single image contained on a given
volume.


* Issue: Many files have bad dark current subtraction.

Lien: The PDS Rings Node will review the images identified as having bad
dark currents and the dark current files used, the dark current
subtraction process, and will examine other calibrated files in an
attempt to identify the cause of the problem. Once the extent and nature
of the problem is identified, the Rings Node will modify the processing
to correct all affected files. The results of this process will be
transmitted to the panel members.


* Issue: The current volume organization, based on the original data
set, promulgates target identification errors from the original data
set.

Lien: The PDS Rings Node will reorganize the volumes, arranging the data
in chronological subdirectories based on FDS number rather than by
target.


* Issue: Many images and the associated entries in the INDEX files
contain incorrect target identification inherited from the previous data
set.

Lien: The PDS Rings Node will correct those target misidentifications in
labels and INDEX files identified by members of the review panel. Any
subsequent corrections will be maintained in an on-line errata site at
the Ring Node.


* Issue: The archive should contain geometry information including
camera pointing and spacecraft trajectory data, and ideally derived
geometry data such as sub-spacecraft longitude.

Lien: The PDS Rings Node will:

(a) Add derived C kernels (pointing information) currently being
produced by the Rings Node and SP kernels (ephemeris and trajectory
information) to the data volumes.

(b) The panel felt geometry information is more useful in the equivalent
of a PDS INDEX.TAB file than in a separate geometry index file. However
a PDS INDEX.TAB file must contain one record for each file on the
volume, and the panel agreed that duplicating the geometry information
six to eight times for each image set was not useful.

The Rings Node will coordinate with the PDS Engineering Node to find a
PDS-compliant approach that will most satisfy this goal. The tentative
plan is to include a minimal PDS INDEX.TAB file on each volume, plus a
supplemental index file that contains one record for each image set.

(c) Add the relevant geometry information to each label file.

(d) At an unspecified later date, the PDS Rings Node will produce an
additional index containing derived geometry information. This will not
be part of the published volume but will be distributed on line.


* Issue: Browse images would be useful.

Lien: The PDS Rings Node will provide browse directories on each volume.
One browse image will be included for each image on the volume (not for
each version of an image set). The browse images will probably be based
on the CLEANED version, although the Rings Node will make the final
determination based on tests of representative image sets.


* Issue: The VICAR headers on processed image files contain misleading
terms and do not always reflect the best available processing factors.

Lien: The PDS Rings Node will improve the documentation (the Rings Node
will not edit the VICAR headers).


* Issue: The ERRATA.TXT files contain very little information.

Lien: The PDS Rings Node will update the ERRATA.TXT to include
information on such issues as known data corruption and the possibility
of erroneous target identification in labels and indices inherited from
the original data set.


* Issue: Additional relevant documents, especially the Voyager Calibration
Report, should be added to the archive.

Lien: The Rings Node will continue to seek permission to include the
Voyager Calibration Report. C. Avis will supply additional documents
which the Rings Node will incorporate (assuming there are no copyright
or other restrictions).


* Additional Issues:

The Rings Node will address all of the smaller comments raised by the
peer review panel and will provide point-by-point responses.