NRAO Home  >  Green Bank  |  Wiki Topic:    GB > Data > BDFITSWorkingGroup

Changes | Index | Contents | Search | Statistics | Jump to Topic

BDFITS Working Group

Welcome. This is the official site for the BDFITS (that's Big Dish) working group.

This page and its related subpages should only be edited by working group members.

The focus of this working group is to unify the efforts of Arecibo and NRAO Green Bank in developing a common format for the storage of data from single-dish radio astronomical observations. The final format should contain all information (in the form of header keywords) necessary for any astronomer to reduce his/her data, and should be in a data format recognized by the international community.

Motivating BDFITS

Arecibo and Green Bank have each utilized the SDFITS single-dish data format. Arecibo has provided data in a format similar to SDFITS for some time, and recently Green Bank has evaluated the utility of SDFITS through a GBT-to-SDFITS exporter.

The primary motivation behind this work is that since the SDFITS format was released, there have been considerable developments related to coordinate axes in general as well as how coordinate axes appear within tables (e.g. the WCS convention). The main functional aspects of these are now described in published standards (viewable through the NASA fits pages http://fits.gsfc.nasa.gov) . The frequency coordinate standard is nearly complete and any changes there are unlikely to be significant. As a result, it is finally the right time to bring the SDFITS convention up to the new WCS standard as described in papers I and II and the nearly complete paper III. Doing so will make it much more likely that readers that know nothing about SDFITS will get at least the basic data and coordinates exactly right. This does not really change the SDFITS convention - it merely brings its use of coordinate systems in a binary table into agreement with the now fully described standard, which wasn't available 15 years ago.

Additionally, the new standard which we have been developing has added a number of keywords which observers have asked for or which have been found extremely helpful. Also, some keyword names have been changed, to bring them into accordance with the keywords' more common usage. Again, we are not significantly changing the SDFITS convention, just updating it to reflect the changes in the FITS format which have occurred over the last 15 years, and adding in information which has been found to be necessary for a complete understanding of the data.

Finally, we are hoping to address in this format a number of the data format issues which have recently been raised, such as how to store contiunuum data which is not well dealt with in the SDFITS format. Since these changes are fairly important differences with the original SDFITS concept, it is important that they be tried out for awhile before anyone considers them set in stone. This is our current goal - to create a new data format which seems to meet all the needs of the community, and then modify it over the next few months until we are satisfied with the product, with the goal of a final product being produced at both GBT and Arecibo by the end of the year.

In sum, the current implementation of SDFITS (see Sdfits) falls short in a just a few areas, and BDFITS seeks to improve upon these shortcomings. In particular:

• The WCS papers have resulted in new, standard, FITS definitions. These definitions are not part of SDFITS. We seek to incorporate these results in BDFITS.
• Continuum data must be stored in SDFITS with one data point per row; it is viewed as one, large, spectral channel. BDFITS will define an efficient means of storing continuum data.
• SDFITS does not contain (in standard form) all header keywords desired by most astronomers.

The work of BDFITS is inpired by the substantial amount of effort and expertise that has gone into SDFITS. Thus, much commonality between the two will be evident.

Timeline

Arecibo Observatory is already using a the bdfits format, as it is outlined here, for the spectral line data being taken with the WAPP backends. Green Back is completing a sdfits writer (http://wiki.gb.nrao.edu/bin/view/Data/Sdfits), and working with the community to insure the bdfits format is acceptable and contains all necessary information for a general observer. Once the format is finalized, work will begin on a beta version of a GBT bdfits writer. The goal (barring unforseen changes) is to have bdfits tested and available for GBT data by the end of 2004.

• Step 1. (8/15/04) Adopt WCS standards and additional keywords for spectral line data.
• Step 2. (10/15/04) Resolve any issues related to continuum data format. GB's initial Sdfits is the starting point.
• Step 3. (12/31/04) Once all issues have been resolved, formalize shared data format between GB and AO.