Date: 2017-03-15

Ludmilla:
 
This is not a very good review; but I found some time and this is what resulted.  You should not expect more; but I will pass along anything additional that comes up in the next 10 days.  Mostly looked at documentation and some labels.  Did not attempt to manipulate any of the data files; was reasonably happy with how that went last time.
 
One important problem: Although this is a raw data set, the documentation includes both the ICD and the REX calibration report, which are known to have significant errors.  In fact it's those errors that have held up certification of all REX calibrated data so far.  We should not certify this raw data set without (1) excising the known errors from the documents or (2) very prominently warning users to avoid those sections.  The invalid material is in nh_rex_radiometer_calib.pdf section (e) and soc_inst_icd.pdf section 12.3.1.2.  

     There is also little acknowledgement that Sides A and B must be calibrated separately; the observations needed to perform the Side B calibration were not collected until last summer and will not be included until at least the next version of the data set.  At best, this version of the data set can only be used for analysis of Side A data and then only if the user locates the calibration observations and develops procedures for applying them to the raw data.  Using the expressions and values in the two documents above will give wrong answers, as documented in previous reviews.
 
Second important problem: The contents of the data set are only vaguely defined.  Since only a subset of Pluto encounter data downlinked through January 2016 is included, there are still important observations that are missing.  dataset.cat says "this data set includes only the Approach data plus a subset of the CORE and departure sequences data that was downlinked through the end of January 2016."  The CORE data include "the bistatic radar THERMSCAN data … [and] the Pluto occultation data for both ingress and egress."  That leaves at least the Charon occultation (ingress and egress) and non-BSR scans (including DISKTHERM) of both targets unaccounted for.  An inventory of all encounter data was compiled for a project review in October 2016, so REX must know what the final data set will look like.  It would be very helpful to users if a summary of that list were included in dataset.cat so that what is included and what is still pending are easy to see.
 
The new files in document/aaa_generic_readme capture some of the miscellaneous material that was scattered (and replicated) across previous archives in a way that makes sense while improving the integrity of the original documents.
 
Acknowledgement that a bistatic radar experiment was conducted is helpful; but the introduction of BSR is conflated with an attempt to describe the radiometry scans (Investigation 3 in rex.cat) and the result is incomprehensible.
 
Annotated copies of aareadme.txt, rex.cat, and dataset.cat are attached.  Most of the edits are for improved clarity; but there are a few lesser errors noted (such as the number of dark sky targets, which is usually one but also given as both one and two in rex.cat).

Dick