Eclipse of Iapetus - May 5, 2007
This page presents my CCD observations of the recent eclipse of Iapetus by the shadow of Saturn and its rings.
Telescope: 20.0cm f/5 Newtonian Reflector
Camera: SBIG ST-7XME
Time Source: WWV
Richmond Heights, Ohio, USA
Latitude: 41d 33' 50.5" N
Longitude: 81d 30' 02.5" W
Datum: NAD 1927
Observer: Robert J. Modic (IOTA)
A CCD time series was performed on Iapetus from roughly 1:46 to 2:46 UT on May 5, 2007. CCDV exposures of 4.0s duration were acquired every 11 to 12s during this time. An exposure of 4.0s was chosen to minimize the saturation of nearby Saturn and yet obtain usable SNR for Iapetus. The time on the computer used for camera control was set to within +/- 1s of UTC using a WWV radio time signal. Additional time keeping drift on the computer may have increased the error in the time stamped onto each image header to +/- 2s over the course of the ~1 hour time series. Sky conditions were good until 2:21, with only very light cirrus present in the FOV. After 2:21, thicker cirrus entered the FOV, making it difficult to see Iapetus on the computer screen. A gradual fading of Iapetus was noted from 1:46 to 2:08. After 2:09, the rate of fading increased until Iapetus disappeared by ~2:22.
Below are crops of three images that show Iapetus at different stages of the eclipse. Titan is on the left. Iapetus is between Titan and Saturn. Dione is to the right of Saturn. Note the bright sky in the last frame due to cirrus. Comparing the relative brightness of Iapetus and Dione, a fading of Iapetus is obvious.
Of over 300 CCD frames acquired, the final 20+ frames were too affected by cirrus to be usable. An additional frame was not used due to bad tracking. The remaining 279 frames were calibrated using master bias, dark and flat frames. Conventional aperture photometry was attempted using two different commercial photometry programs. The brightness of Saturn made it too difficult for these programs to get an accurate measure of the sky background around Iapetus and the resulting measures were too poor to be useful. I then used a program called Extract, written by Anthony Mallama for reducing CCD observations of Jovian satellite eclipses. This software rejects the highest and lowest 25% of pixel values in the sky annulus before computing the sky background. Extract also uses offset tracking relative to the brighter comparison star (or moon) to follow the eclipsed object even after it fades from view.
After learning to use Extract properly, I was able to perform differential aperture photometry on all 279 frames successfully. Titan was used as the comparison "star" and Dione was the check "star". The results are presented below.
The plot above shows the ratio of the sky subtracted fluxes of Iapetus/Titan plotted vs. UT. An aperture of 5 pixels diameter was used for this plot. Below is a similar plot of Dione, again using Titan as the comparison object and an aperture of 5 pixels.
Since Dione was located about the same distance from Saturn as Iapetus, but on the opposite side of the planet, it should be a good test of the accuracy of the photometry software. During the 1 hour of time series, Dione would not be expected to vary much in brightness. A plot of Dione should then be essentially a straight line. The plot above does indeed show a constant brightness for Dione, with increasing scatter due to cirrus towards the end of the time series. This result gives confidence that the variation in the brightness of Iapetus shown in the first plot is real.
The plot for Iapetus does show an asymmetry. There is a very gradual fade from 1:46:37 to 2:09 UT, after which the rate of fading increases. The likely explanation for this is that Iapetus grazed the shadow of Saturn's "A" ring before being eclipsed by the planet's shadow.
I would welcome the collaboration of any other observers who recorded this eclipse. A combined analysis of all data from this event might help untangle the affect of the rings' shadow on the eclipse timings and allow a good refinement of Iapetus' orbit. I can be reached at rjmodicATn2net.net.