Forums / The Science / Photometry / Good results using Intermediate Calibration spreadsheet
Good results using Intermediate Calibration spreadsheet
Hi all,
It's been a while, but over Thanksgiving weekend I decided to try out the intermediate spreadsheet and take some images of the Epsilon Aur area to see how it would match up. After crunching all the numbers and looking at the results, I'm very happy with the precision I seem to be getting. Many thanks for creating the spreadsheet which made the process a LOT less time consuming than if I had to do it myself.
To obtain magnitude measurements, I used my modified Canon XT camera and a 50mm lens. Each measurement has a series of 5 star images (with flat frames and dark frames), aligned and stacked. The camera was out of focus enough so that each star covered about 15-20 pixels in diameter. For each measurement series, I calculated an Extinction and Transformation Coefficient. I took the images at a time when all the stars measured had an airmass of 1.2-1.3
I'm attaching a spreadsheet which summarizes my results so far. I do have a couple of questions though:
- My extinction coefficient varied by quite a lot (-0.03 to -0.259). All images were taken within a few minutes of each other, so I was not expecting the such a wide range. My average ended up as -0.116 with a standard deviation of 0.072. Does this seem normal?
-After all my results, I am wondering about the listed magnitude for Omega Aur. All of my other check stars come within 0.01 magnitude or less of the published value. My average result for Omega Aur differed by 0.023. Granted that's still pretty close, but I've see a couple of other sources that give the magnitude of Omega Aur as 4.93 or 4.94 (a better match to my results) rather than 4.95. Has anyone else had a similar experience? SIMBAD gives a V magnitude for this star as 4.989 and also mentions an 8th magnitude companion, so I wonder if adding the two together are causing different results?
==
Mike Durkin
Hi Tom, Honestly, Idon't do the calculation myself, Ilet the calibration spreadsheet do the real work: http://www.citizensky.org/sites/default/files/DSLRPhotometry/spreadsheet... Iwould image that the procedure you use would work just as well with a DSLR. Your comment about taking images during short period of time may be relavent in this case, all of my images were taken during a 5 minute window. Ithink that I'll try your suggestion and measure some star magnitudes as their airmass changes and see what Iget. -Mike Durkin
Hi Mike, On behalf of the DSLR team: you are most welcome for the spreadsheet. We were hoping this would prove to be quite useful and, eventually, gain broader adoption. As for you question about the extinction coefficient, I'm afraid your results will be typical. Due to various atmospheric effects the extinction coefficient, as well as the other parameters will drift around quite a bit. This problem is further complicated by the least-squares fit which happily alters parameters as needed to fit the data. How do the residuals to the fit look? Do they fall in place well? In generalI would suggest doing what you are already doing: calculate the averages of each star over several frames and use that value instead of the individual frames. An quick way to improve your extinction determination is to include stars further out. You might see if the II/168 catalog has a better set of calibrators elsewhere near eps Aur, especially to the North East. As for the catalog value for Omega Aur., the catalog quoted on the website came with recommendations from Arne. The uncertainty in the V-mag is 0.022 and the error in B-V color is 0.01 mag so the values you are getting are acceptable (they surely fit within the error bars). Note that if we were only using one comp star you couldn't legitimately claim an uncertainty lower than the highest uncertainty from a calibrator. Using the least-squares fit and some cool statistics we can formally push the uncertainty down to the single-millimag level. Now we haven't discussed this in much detail, but I hope to formalize the mathematics in a publication sometime soon. Keep up the good work! Brian
Mike: How are you determining your extinction coefficients? I am doing photoelectric photometry and determine my extinction coefficients in the following way (a DSLR might require a different procedure, but I think this would also work): I measure a standard star as it changes airmass for a period of about one hour (this is about how long it takes to get one PEP data point). When the brightness of the star is plotted against its airmass, the slope of the resulting line is the extinction coefficient (the units are mag. per airmass). I run into trouble when the star is almost overhead and the airmass is not changing much-- any variation in atmospheric conditions during the data run will produce all sorts of wild slopes that aren't accurate. I imagine the same problem might occur if all of your measurements occured during a short period of time-- a minor change in extinction due to some sort of atmospherics would look like a major change. Try taking images every 5 minutes for half an hour or so-- this should give you 5-6 data points and so give you a good-looking line. Tom Rutherford