For Data Preparers

______________

Preparing data for submission to the PDS includes not only labelling of data files and collection of documentation, but also content verification and usually at least some measure of reformatting to get the data into one of the standard deep archive formats required by the PDS. The formatting and documentation requirements of the PDS were developed in conjunction with astronomers to ensure that the data archived will be accessible long past the lifetime of the mission, its collaborators, or its hardware/software installation. The PDS format is specifically designed to be an archive format, not a working data format. Listed on this page are pointers to information and software tools to assist you in preparing and verifying data for submission to the PDS.

______________

Introductions

The full PDS archiving system and standards can seem very imposing when first encountered. Several more user-friendly introductions have been prepared to help new users get up to speed quickly. All of these introductions are available from the Engineering Node of the PDS.

______________

A Word About Data Formats

It has been our experience that simple data formats have several major advantages over more complex or proprietary formats:

  1. They are immediately accessible to a much wider section of the science community.
  2. They are stable over decades, maintaining accessibility for the projected life of the archive.
  3. They are easier to document and verify, facilitating efficient review and archving.

Therefore, the Small Bodies Node is a strong advocate of simple data structures, especially in high-level mission data products. The simple data structures are homogeneous arrays and flat (ASCII or binary) tables. Not surprisingly, these are also the formats well supported by the Flexible Image Transport System (FITS) standard, which provides a very general and widely accessible format for exchanging file between various platforms.

Note that, from an archiving point of view, headers (and trailers) that are appended to the data structures present no major problems, as long as they do not interrupt the byte stream which constitutes the observational data. Thus any compliant FITS file can be easily prepared for PDS archiving without modifying the original FITS file because the headers occur before the individual data structures.

______________

PDS Standards Documents

In addition to the printed documents available from the Engineering Node of the PDS, there are on-line versions accessible by Web browser:

Data Preparation Workbook

This workbook provides a quick overview of the PDS system and what constitutes an archive dataset, then walks you through the process of designing and creating and archive.

Archive Preparation Guide

This workbook provides a cook-book approach to preparing data for submission to the PDS. constitutes

Proposer's Archive Guide

This workbook describes the information needed to prepare a proposal. This document includes examples and templates.

PDS Standards Reference

This reference contains the detailed PDS standards for things like archive layout, data label requirements, units of measure, time values and other aspects of the PDS archives.

Plantery Science Data Dictionary (PSDD)

The PSDD contains the definitions for all keywords and data structures used in and available for archive datasets. A flat ASCII version of the PSDD is required to run some of the software tools; there is a Web interface into the dictionary which allows you to retrieve specific keyword definitions.

______________

PDS Tools

A number of tools have been developed for generating and verifying PDS labels. The complete inventory of general PDS utilities is available from Engineering Node. The tools our data preparers have found most useful are:

______________

SBN Tools

The SBN has also developed suites of small utilities for working with the the types of data we most frequently encounter, and to produce formats that are preferred by the SBN (which may be more restrictive in some respects than the general PDS requirements). The majority of these routines are written in Perl; some are written in ANSI-standard C. All were developed and tested on a Sun Solaris work station. The source code for these routines is available from the SBN, as is, from our software archives. Utilities available include:

If you need some specific tool for preparation of your archive and could not find it among our tools, please, contact Anne Raugh (email to raugh at astro.umd.edu). It may be that we already have developed the tool you need, but it is not online yet. We also can develop a special tool upon your request.

Stylesheet

The SBN Style Sheet for PDS labels and ODL files provides notes, reminders and examples regarding elements of style expected in archive product labels and catalog files submitting to the SBN for archiving.

 

TARGET_NAME Conventions

______________

Before Submitting Data

Before submitting your data to the SBN, please run the following verifications:

In addition, if you are submitting data in tabular form (TABLE, SPECTRUM or TIME_SERIES objects), you should also run the following utilities:

In all cases, if some of the labels were generated by the same label-writing program and were not subject to individual editing, then it is sufficient to run the verifiers on a representative sample. All files in the series must be corrected, of course, prior to submission.


Please do not submit data to the SBN that does not
pass at least this minimum format verification.


______________