Abstract
Live images of the Total Lunar Eclipse of 1997 September 16 were successfully transmitted to the Internet from the Calwell Lunar Observatory. This resulted in tremendous interest from browsers of my live images on the web and the three animations placed onto my Home Page. These included the low-resolution live images with cameos from Sweden during totality, another from my high-resolution images and a third sequence made from images e-mailed from Switzerland.
Two CCD cameras were mounted on the telescope which was remote controlled from the computer to centralise all of the captured images of the eclipsed moon. Selected images were frame grabbed by computer, saved and processed to a web compatible format before they were transferred in a time sequence to the Internet web page by direct file transfer protocol. After the live transfer was concluded, animated GIF files were added to my web pages.
Later, calibration of the eclipse frames by image analysis techniques allowed the semi-diameter of the umbra to be measured to determine dynamic change in the umbral geometry. Highlights from the large number of e-mail messages received are included and the background to this live world wide web project with links to my Internet pages are given.
Introduction
The very successful live capture of images of the moon during the recent total lunar eclipse of 1997 September 16 are discussed. The weather in the Canberra area was superb with only five minutes of cloud during the complete eclipse. Two CCD cameras recorded the complete eclipse on video tape from first contact to moon set. Sequential images were transmitted live to the Internet from a low resolution colour camera while a high resolution monochrome camera also recorded images. These images have been calibrated and measured to determine the geometry of the umbra.
Animations were prepared from both sets of images which were placed onto my Internet Home Page with colour images from Sweden included in one sequence to cover totality. A third animation from coloured images received from Switzerland was also included in the web page.
The live eclipse project proved to be highly successful and generated considerable amount of e-mail concerning the live images of the eclipse and later about the animations. In addition, many crater timing observations were received for analysis.
Background
After observing and promoting interest in most lunar eclipses since 1972, the live imaging project proved to be a zenith in this amateur research. Computer programs were first commenced in 1984 to produce data for lunar eclipses and to establish a technique for the prediction and analysis of crater timings. Later statistical methods were developed to determine the umbral geometry, in particular techniques to estimate the umbral oblateness, or flattening.
In 1991 the first video recording of a lunar eclipse was undertaken with image analysis of these records by measurement of the calibrated umbra to determine if dynamic change in its geometry could be detected. Then following a co-operative program with Brazil in 1995, the concept of placing LIVE images onto the Internet was first considered. With my low budget program these recent eclipse images were captured, saved and put to the Internet by very direct methods when compared to the elaborate software used in Brazil.
However, successful imaging was achieved thus making 1997 September 16 an appropriate highlight for my thirtieth lunar eclipse in the Soulsby Series.
Image Capture
The instrumentation used for image capture is illustrated in Figure 1 and consisted of two refractor telescopes both fitted with CCD cameras. The telescopes were attached to a quartz driven equatorial mounting. One instrument was a 400 mm telephoto lens, with a 20 mm eyepiece focussed into a Sony Handycam colour CCD video camera with its time generator adjusted to UT. The other instrument was a 40 mm aperture Unitron with a 12.5 mm eyepiece focussed into a Mintron 650 line monochrome CCD surveillance camera.
Both instruments were fitted with lens hoods, UV and yellow filters to reduce full moon glare with the Unitron eyepiece also fitted with a lunar filter. The electronic shutter of the Sony Handycam was used to further reduce glare and the Unitron was masked to 50 % aperture from full moon to second contact. As it would have been difficult to remove and refit the filters while at the same time, transferring images to the Internet, neither camera was sensitive enough to record totality or stellar occultations.
Telescope remote control
The relatively light weight twin telescope instrumentation replaced my 15 cm Newtonian to achieve the required low magnification to give a complete image of the moon in each field and to enable satisfactory remote control of both axes of the equatorial mounting.
The right ascension axis was quartz driven and easily adjusted by cable from the observatory to the computer to maintain a centred image of the moon in the vignetted field of the Sony Handycam. The declination axis was also remotely controlled by cable to a 6.3 volt DC 200:1 geared motor. The control proved to be quite sensitive to the inertia and balance of the two mounted telescopes and CCD cameras, but following very exact balancing, good drift correction in both declination directions was established. The moon was tracked to moon-set over the Brindabella Mountains with only minor interference from cloud and from one small tree near the horizon.
Web transfer
As images were captured with my QuickImage 24 frame grabber and Macintosh IIci computer they were saved for later analysis, processed from *.PICT files to *.GIF files using Graphic Converter and "put" to my Home Page using the File Transfer Protocol (ftp) software Anarchie.
Image Analysis Methods
Each captured CCD image was examined using Image Analyst, a commercial software package ideally suited for the calibration of the moon and measurement of the umbra. One such image result is shown in Figure 2 where the calibration of the moon is shown in the Measure box with the true moon's semi-diameter of 0.2789 degrees at the image time of 17h 34m 13s (UT). The measurement of the umbral semi-diameter is by a best-fit-circle placed at the maximum rate of change of brightness at the umbra edge using a greyscale of 50 which can be adjusted in the Histogram box.
Umbral change
Measurements of umbral semi-diameter during the lunar eclipse are compared with the topocentric umbral geometry (Soulsby, 1996) which is computed for the location of each observatory.
A good result for the mean change in umbral size was obtained by comparing the measured images from the Calwell Lunar Observatory high resolution Mintron CCD images with the computed topocentric umbral geometry as shown in Figure 3. These images were analysed at 3 minute intervals with a break at totality. The computed topocentric umbra gave a typical smooth result, while the departure of each measured image was a little scattered, it produced a mean increase in umbral geometry from 0.15 to 0.35 degrees over the time of the eclipse.
Crater timing analysis
To date, seventeen observers have reported crater timings from this eclipse. Five of the astronomers were from Australia, two from Spain, one from Italy, two from Germany, six from Slovakia and one from England. Their crater timings have been analysed to give the umbral enlargement, and for those that meet the statistical criteria, the oblateness of the umbra. A typical oblateness estimate from a best fit linear analysis is shown in Figure 4.
The crater timing reductions gave an overall mean umbral enlargement of 2.1298 +/- 0.1100 % from the 328 crater immersion timings and 1.9476 +/- 0.1420 % for the 212 emersions, both values very near the expected 2 % umbral enlargement. The more subjective oblateness estimates, computed from a best-fit-ellipse routine, varied from a maximum reciprocal value for crater immersions of 1/36 (or 0.0313) to a minimum of 1/284 (or 0.0035). Surprisingly close values of 1/32 to 1/185 for crater emersions were also computed. The expected geometric value for this particular eclipse was 1/221 (or 0.0045), so the umbral oblateness was greater than that expected by factors as high as 7, to a low of 1.4. Tabulations of these reductions can be found on my Internet Home Page which will be up-dated as further observations are received.
It was pleasing to find that the previous mean value for umbral oblateness of 1/102 lies in the mid range of the values determined from this eclipse.
e-mail Highlights
The number of web page accesses during the eclipse period advised by my ISP are shown in a chart of the number of "hits" below:
Here are just a very few of the hundreds of e-mail queries and reports received, starting with the most recent on 1997 October 3 from Egypt.
From Cairo Egypt:
Thanks for your recent info about the eclipse time in Cairo. We had a wonderful view sitting on the roof of our building, 13 stories up. A number of the neighbourhood kids also joined us and I passed your info on to some of the teachers in my school. None of these people knew about it beforehand as there's not a lot of information about such things in the local press. The time was exactly right, though an hour later (20.08) due to Egyptian summer time being 3 hours ahead of GMT, but I knew that. Summer time here runs from the last Thursday night in April to the last Thursday night in September, generally - nothing is sure in Egypt. (A few years ago, Ramadan was in May, so we changed to summer time as normal, then winter time again for Ramadan, so people could eat at an earlier time in the evening, then we changed back to summer time when Ramadan ended. It was very confusing.) We spent about 2 hours watching the eclipse then the kids had to go in for they had school next day, but we kept looking out till it finished. Thanks again for giving us the chance to see such a beautiful sight.
Eva Abdin
From Brasil:
Congratulations on your live eclipse coverage. We watched many of your images of the eclipse. Yours was the only WWW site which showed so many pictures of the event. I had notified many colleagues of your transmission so that I think many of them also monitored your WWW pages. I read it was a bright eclipse as expected.
Greetings from Brazil. Helio C. Vital (vitalhc@centroin.com.br).
From Sweden:
I just thought that you might want to know about my new page with almost 20 pictures of the lunar eclipse that I managed to take between the clouds!
Best regards from the other side of the world!
Bengt Ask, (bengt@df.lth.se)
From China:
I just wanted to express our appreciation for your information and that of our 6 year old who had a great time looking. Here in Beijing, we are not really set up to take measurements but you might want to check with the Beijing observatory since the seeing was about as good as it gets here. A front came out of Mongolia the night before with moderate winds and blew all of the smoke and dust out of the air, dropped the humidity and the temperature and generally made things bright and clear. If you didn't know, this eclipse occurred on the night of the Chinese mid-autumn festival where people eat moon-cakes and fruit and watch the moon.
Best Regards, Charles Krabek.
From Athens National
Observatory:
Congratulations for your fantastic live coverage of the lunar eclipse.
Bad weather conditions did not allow us in Athens to see this phenomenon directly. Thanks to your web page, many enthusiasts that visited our Institute during the last night were impressed by the indirect live imaging of the eclipse.
Best regards, Dr Ioannis Bellas.
Reference
Soulsby, B.W., "Topocentric umbra", Australian Journal of Astronomy, March 1997, 7(1):16-22(1997).