Calibration Upgrade Team: John Benson, Ed Fomalont, Greg Taylor, Craig Walker, and Joan Wrobel with help from Alison Peck, Tony Beasley, the Goddard VLBI group, and others. History: v1.4 4/27/2001 include comments from Craig and Steve v1.3 4/18/2001 include comments from meeting (Claire, Joan, Ed, Craig & John) v1.2 4/17/2001 include comments from Ed v1.1 4/9/2001 include more comments from Craig v1.0 4/6/2001 include some comments from Craig v0.1 4/4/2001 original draft by Greg Taylor Draft v1.4 4/27/2001 Specifications for a Calibrator Search and Evaluation System A fundamental need of VLA and VLBA users is to search out the best calibrators for a given observing program. At the present date there are two major databases (with about 800 sources in common) and a large number of tools in various states of disrepair. No common interface is presented to the user, and even worse is the fact that most of the tools are disjoint. For example there is a tool to search the VLA calibrator manual, and another tool to look at the flux history. Even after manually passing information from one to another the user frequently needs to consult the VLA calibrator manual to find out positional accuracies and to look at images and visbility plots. The situation for VLBA calibration is perhaps worse. Information about calibrators is even more widely dispersed and the search engine is incomplete, inadequate and irreperable (the code has been lost). To remedy the current situation we would like to establish a single database for both VLA and VLBA calibrators, and a suite of tools to access it. Below we attempt to describe the desirable specifications for such a system from a users point of view. Conceptually, we'll consider three steps for the process: (1) Search; (2) Evaluate; (3) Include. Lastly we provide a few guiding remarks about (4) Infrastructure. 1. Search In this first stage the user has in mind a specific telescope and frequency and a target RA, Dec or source name and desires to find suitable calibrators in the vicinity. Generally the angular seperation between calibrator and target is an important consideration which should be available both graphically and numerically. The current tools in use for the VLA and VLBA are both clumsy. A much better interface has been developed by James Morgan for BIMA (see http://bima2.astro.uiuc.edu/java/Calfind/). This interface allows the user to input: Field Size in degrees Planet Epoch + update button center RA, DEC + update button What it provides is a RA vs Dec skyplot with the target location and that of calibrators. The sun and planet are also plotted for the planet epoch specified. The calibrators are color and size coded for brightness. A simple uptime display is also shown in a panel to the right of the skyplot. If a calibrator is "clicked on" then it returns one line of information about it in a text window at the bottom, including the most recent flux density measurement, and the source-calibrator angular seperation. This is an excellent start to the functionality we would like. I sent e-mail to James Morgan on March 21st asking him if he would be willing to provide us with his JAVA applet and he has delivered it to me. 2) Evaluation. The BIMA Java applet provides some information about the calibrators, but not as much as VLA and VLBA users need to make an informed decision. So even if we could begin with the BIMA calibrator finder as a starting point, there are still a number of features we would like. These additional features are: A) A frequency, and/or flux density, and/or angular size filter such that only calibrators meeting certain criteria are shown. This pre-selection might be necessary to avoid users being swamped with a large number of nearby but faint/resolved choices. The best way to implement this would be an interface on the main plotting widow allowing the user to select such items as: (1) desired coordinate frame (J2000 or B1950); (2) Frequencies of interest (perhaps by using buttons to select bands); (3) the instrument of interest (VLA/VLBA/ALMA -- each of which could select what information is returned about each calibrator like the default visibility plot, and would add some features like latitude lines on the plot for the selected instrument; and (4) a flux density cutoff. B) Links to images and visibility plots. Ideally these would appear in a compact one-line form when one clicked on a calibrator. C) A link to a light curve from the VLA flux history database (http://aips2.nrao.edu/vla/calflux.html). D) Link to a spectrum (total flux vs frequency) using most current measurements but also with historical ranges plotted. E) Additional textual information about the calibrator like positional accuracy, (u,v) restrictions. This may require a couple lines like: 0319+415 code=A dist=2.23 deg Aliases J0319+4130 0316+413 3C84 J0319+41 J2000 Error gal lat/long Date (4 Mar 1997) RA 03 19 48.160500 0.0054 -13.26115 03 19 49.123456 Dec 41 30 42.10300 0.00032 150.57583 41 30 04.90514 Position from JVAS - Patnaik et al. 1992, mnras, 254, 655. Comment: Good VLA calibrator, heavily resolved with VLBI. Polaration Calibrator (link to polarization monitoring data) BAND VLBA A+ A B C D S_T(Jy) S_short S_long Date MIN(kL) MAX(kL) =========================================================================== 90cm P X S S X X X 8.0 3.0 0.8 1996Feb01 13 50cm X S - - - - 15.0 5.0 1.2 1996Feb01 13 20cm L X P P P X X 23.9 6.0 1.3 1996Feb01 13 13cm X P - - - - 23.5 7.0 1.4 1996Feb01 6cm C X P P P P P 23.3 8.0 1.5 1996Feb01 3.7cm X X P P P P P 21.7 9.0 1.6 1997Mar12 2cm U X P P P P P 20.7 10.0 1.7 1996Feb01 1.3cm K X P X S S S 16.4 12.0 1.8 1996Feb01 7mm X P X X S S 13.0 10.0 1.9 1996Feb01 3mm X P - - - - 7.0 5.0 1.2 1996Feb01 1mm X ?? The band codes indicate default continuum frequencies commonly used. If other, intermediate frequencies become commonly used by the EVLA then they could warrant additional lines. Default plots obtained by clicking on a calibrator: ---------------------------------------------------- | Plot | | | log Flux | Show latest and historical range. | | | | | ---------------------------------------------------- 75 MHz Frequency 100 GHz ---------------------------------------------------- | Plot | | | Flux | Show visibility data (well averaged) | linear | at default frequency (or closest available) | | most relevant for the instrument selected | | (VLA/VLBA/ALMA) | | | ---------------------------------------------------- 25m log(Baseline length) 9000km Click on quality code for links to visibility curve, image, flux history, spectra, and other information. Clicking on one of the links should pop-up a postscript or gif image in a seperate window. This textual information (and links) could appear in a special box that updates each time a calibrator is selected in the main RA-DEC plot window. Some of the links could also appear in a drop-down style menu so as to provide the maximum amount of information with the minimum amount of clutter. 3. Include After finding a calibrator the user should be able to mark it for inclusion in a VLA (jobserve) or VLBA (sched) file. At the end of a session the user should be able to take a file containing selected calibrators into a schedule. 4. Infrastructure Most of this document up to this point is comprised of discussion about a suite of web-based tools to access the unified calibrator database. In this section some demands on the database itself are discussed. A single calibrator database should be established. This would help to keep the VLA and VLBA in the same reference frame. It would also reduce our overhead costs of maintaining seperate databases and tools. Authorized people could check in additions and corrections to the database. Epochs should be attatched to all additions and corrections to provide a history for those that need it. Only J2000 positions would actually be stored and other coordinates would be derived on the fly. Although most people will query the database through the web-based tools described in section 2, it may be that other programs may want to dip into the database and retrieve information. Jobserve and Sched, to name the two most popular programs for creating VLA and VLBA schedules, may want to build in some functions to either interact with the web-based tools, or to retrieve information from the database directly. Other programs like cjobgen will also make use of the database. The ALMA observing tool may require the ability to query the calibration database (through some TAQL string or special function) be the scheduling software. Or people may want to write their own special-purpose scheduling software that uses the calibrator database. --- end of specs