next up previous contents
Next: Simple Quick Look Analysis Up: OBSERVING version 1.2 Previous: Display Words

Seeing Measurements

SEEING and SPECSEE


A seeing routine for images has been written that fits a Gaussian to the sky subtracted intensity profile of a star. The routine will fail (gracefully) if the star is too close to the edge of the frame or the mean sky has a negative number of counts. It will fail somewhat less gracefully if the seeing is terrible (full width at half maximum exceeds about 13 pixels). It doesn't work too well if the seeing is very good, and the data is too undersampled (i.e. full width at half maximum less than 1.5 pixels). The word functions once an image has been displayed on the monitor. When you type SEEING, the cursor is turned on. Move it to the center of an appropriate isolated bright star, and hit the space bar. SEEING will then give the FWHM in pixels and arc-sec of the stellar image for a cut through the center of the stellar image in both x and y directions. This can be repeated as many times as desired. Type Q to quit.

SPECSEE does the same thing for spectra, fitting only in the one dimension perpendicular to the dispersion. This routine will fail if the seeing exceeds 15 pixels. SPECSEE begins by asking if the display is from IMAGE or from SIMAGE, after which it continues in a manner analogous to SEEING.


Seeing Log


To help in our efforts to improve the seeing and to monitor whether a given change produces any effect, a seeing log file has been created. After each measurement with SEEING or SPECSEE that appears valid, you should allow the measurement to be logged by answering yes to the appropriate question. The value written to the seeing log is the average of the seeing deduced from the x and y cuts, so be sure that both values are reasonable before allowing a measurement to be saved in the log file.

The seeing log file is in the system disk, not in the scratch disk. In order for the seeing measurements to be correctly converted from pixels to arc-seconds, please be sure that the image scale is correctly set, using the word SCALE, even for spectroscopic frames. Also if you change the binning, please reset the scale appropriately.


next up previous contents
Next: Simple Quick Look Analysis Up: OBSERVING version 1.2 Previous: Display Words
Patrick Shopbell
7/2/1998