Let's suppose that you and a partner have recorded several hours of lunar video from different locations. How do you reduce it? Slowly, meticulously, and (on the first pass) independently. You are looking for a ^mag-5 "flash" that covers only a few pixels, and which lasts only a few video frames. Played back at "normal" speed, this flash may be very hard to detect. I have watched videos of the 1999 Leonid flashes, and even after being told in advance where in the image the flash would occur, and at what time, I missed them on the first showing. A casual viewer is not likely to notice these subtle but important flashes! Plan on going through your videos several times at "normal" speed. This can be done visually, with patience and attention. That is the approach that most observers use. If you have the option of playing the record back at a slow speed, your ability to detect the flash should improve.
Some observers are experimenting with frame grabbers and digital image-processing methods to pre-screen the video, (hopefully) simplify the detection process, and (again hopefully) improve the odds of detecting flashes that are embedded in the video. If these developments are successful, you'll see reports on the ALPO website. If you are experienced with this technology, you should definitely consider bringing it to bear on your video data.
Each candidate flash must be examined carefully to see if it betrays the distinguishing marks of a false alarm vs. the characteristics of a genuine "impact flash''. The most common false alarms are cosmic-ray point flashes on the video sensor. These tend to be single-frame events, and may be either single-pixel or oddly-shaped extended artifacts in the image. A good candidate impact flash will last two or more video fields (i.e., it's not a single-video-field event), will show a star-like point spread function (i.e., it's not a single-pixel event, and it's not the sort of odd-shaped thing that you commonly record from cosmic-ray hits). For each candidate event, record the time, duration, and location on the Moon. Use the images of your "comparison stars'' to estimate the brightness of the flash (in its brightest frame), and record its brightness along with any other distinguishing features.
After each observer has examined his or her tape, they should compare the times of any detected flashes, and their location on the Moon. Flashes that appear on one record but not the other should be examined carefully on both video records. It may be that one observer simply missed it during the initial examination of the video. A flash that is found at the same time, and appears in the same location on the lunar surface, on both records, is a candidate lunar impact. The criteria for confirmation are quite stringent. The times of the two detections must match within ±2 seconds, and their locations on the Moon must match within 2 degrees in latitude and longitude in order to be considered a "confirmed" detection.
Both "lone observers'' and local teams should submit the complete record of their observations to the ALPO, so that observations from many sites can be combined, compared, and permanently recorded. The ALPO Lunar Meteorite Impact Search project is coordinated by Mr. Brian Cudnik. The project's website is located at: http://www.lpl.arizona.edu/~rhill/alpo/lunarstuff/lunimpacts.html
Your report to the ALPO coordinator should consist of two parts: a data sheet and a lunar map. The format for the data sheet is shown in Figure 6.3. It is self-explanatory. You must report who you are, where you observe from, your observing method ("video recording''), the times during which you monitored the Moon's surface, sky conditions (seeing, transparency, and clouds). For the video-recording method described here, you do not need to worry about "reaction time''. Then, list all the events you observed. For each event, include the UT date and time, duration and magnitude of the event, and any other comments or unique features of the observation. (If you're not familiar with Universal Time, see Appendix A.)
The second part of your report is a map of the lunar surface, showing the location of the terminator, and of your observed flashes. An example of this is shown in Figure 6.4. Note that it is critical that you accurately cross-reference each "flash location'' on your map with the corresponding "event #'' on your data table.
If you have organized a local team, each observer's data should be separately reported to the ALPO, so that their database is kept accurate. For flashes that were confirmed by another member of your "local team'', note the confirmation in the "comments" column of your report.
Because part of the value of this project consists of gathering data that will give a better understanding of the statistics of lunar impacts and the associated flashes, it is important that you record in your notebook, and report to the ALPO, all of your observing sessions, even if some of them do not yield any observed or suspected flashes. This way, the approximate rate of flashes can be determined. If everyone were to report only their "successful" observations, then this important rate information would be lost.
As in other projects, be sure to archive the tapes (or digital copies) that show impact flashes—they may be requested by astronomers doing further analysis of the events.
Was this article helpful?