The Simbad/AO Tool

For most observing at Keck we require little information about the observing targets than their coordinates (RA and Dec.). For AO, however, we often are asked to supply some more information, namely the brightness and color of the star. These are parameterized as the V magnitude and the B-V color. The latter is simply the difference of the B and V magnitudes.

The reason for the V magnitude is to allow a reasonable estimate of the expected count rate from the star. This gives the auto frame rate tool a good starting point from which to estimate the AO parameters.

The B-V color is used in the DAR (Differential Atmospheric Refraction) correction software. While the DAR correction software is not always implemented, we are trying to get to the stage where we used it all the time behind AO. The idea behind DAR is that the atmosphere bends starlight by differing amounts depending on the wavelength of the light, and the elevation of the source. This "differential atmospheric refraction" always occurs in the elevation direction, with blue light appearing to come from higher in the sky than red light. What this means for AO is that, while the wavefront sensor keeps the optical image of the star fixed with respect to the AO and science optics, the infrared image on the science detector will move depending on the color of the infrared light reaching the detector, and the elevation of the object. The elevation is known, and the amount of offset between infrared and optical images is a function of the infrared wavelength (set by the NIRC-2 software, for NIRC-2) and the effective wavelength of the wavefront sensor. The effective wavelength can differ from the central or mean wavelength because different stars will have different weighting on the blue and red ends of the wavefront sensor bandpass. This has been calibrated as a function of B-V color, using various stars, and hence the need for B-V in order to accurately predict this parameter.

How do we get V and B-V?

There are currently a number of options:

• Look in the starlist.
  If the observer has placed the information into their starlist, you can read it from there, using a text editor or listing the starlist on an xterm. Unfortunately, SKY does not currently handle this task, so even if the information is there, you need to enter it by hand.

• Ask the observer.
  If the observer does not know the parameters, you may have to resort to the next method...

• Look up the information from some database.
  Currently we do not have good databases for this, but the on-line Simbad database contains a lot of this information. Unfortunately, it is cumbersome to find your browser, find Simbad, find the proper data entry screen, type in the star name, then look through the results, find V and B, and subtract the two to get B-V. This has led to...

• Simbad.tcl
  This tool has an automatic and a manual mode. They are combined now into a single GUI, where you can turn off the auto-update feature if you like. The Simbad.tcl description follows.

Enter Simbad.tcl

First note the button labeled "autoload target" below the Target: field. When this is highlighted (as shown), every time you change the TARGNAME keyword in DCS the new value will load into the "Target" name field, and the tool will go out to Simbad to try to get the B and V information. If you click the button to turn off this feature, you will have to type in the target name by hand.

Once the target name has been entered, even if it is via the autoload feature, you can edit it. This might be necessary if, for example, the starlists contains an unorthodox name for the target, as in "HDC 12345", or contains numbers preceding the name, as in "24. HD 12345".

After typing in a new target name or editing a name, either hit or click the "Get info" button. The "Get info" button will change to saying "Working..." if it is out trying to contact Simbad and retrieve the information. When it succeeds, the button label will change back to "Get info" and the new numbers will appear.

If Simbad is unable to identify the star, or if it simply does not contain the relevant information, you will get question marks instead of numbers, e.g. "V=? B-V=?". Sometimes, if the V magnitude is available but B is not, you will get only a V, and B-V will have a question mark. This is fine for the autoframe rate tool, but not for DAR correction.

Once you have the information, you can set the AO keywords (which are TARGVMAG and TARGBVMAG) by clicking the "Set" button. This will rsh onto ulua to set the AO keywords.

The goal is to have this be as highly automated as possible. If, for example, we run into a lot of starlists that have preceding numbers ("24. HD 12345"), we can try to strip off any such numbers automatically. We have to beware of not corrupting legitimate star names like "24 Peg", however. Feedback on problems encountered is appreciated to help fine tune this tool.

Bob Goodrich
22 November 2002