This event goes through Watsonville, with the northern limit on the north side of Santa Cruz. But the rank is low enough that there's no great advantage to drive to Watsonville. The centerline odds are only 47%, and 41% from home. Karl is north of the northern limit, but well within the 1-sigma line, and decent odds of about 25%. The star is fairly bright, at a good 56 degree altitude and on the right side of the Tail of Cetus, Az=227.
|
|
I drove to Mar Monte Rd and set up at the end of the cul-de-sac and used Polaris and Hamal as my align stars. It went to the target to within about 3/4 of a Q70 field of view. Seeing was not great, and lots of scintillation. Kirk observed from home and also got a solid event at 2x setting. Karl tried it too, but looks like the tape as the dreaded "bars". Will they disappear when the tape heads are cleaned and the camera warms up? We shall see...
One neighbor came out, with her dog, asking what I was doing, and was satisfied I was an astronomer. Said there's been "wierd stuff" going on lately, including a shooting (?). Anway, no such dramatics for me. The skies looked clear, although I see evidence in the tracking star of perhaps some very thin cloud going by. There was ~2 sec oscillations in the tracking star, and so I used it as a reference for the very faint target star. I was surprised the target was as hard to follow as it was. I went with 4x setting. I think the focus and scintillation was not as favorable as it was the previous night for Ahrensa. Still, it's a very solid occultation positive. The duration is longer than expected: 3.4 seconds vs. 2.3 seconds predicted. Oblong perhaps? Very dark? I was surprised that the asteroid wasn't easier to see. It did not look like a 12.2 magnitude object, it looked more like 12.8 as far its visibility.
From the IOTA map, it looks like the velocity of the shadow was close to about 500 miles/min, or 8 miles/second. I was about 12 miles east, or later, along the path, so to correct my timings to be in line with Kirk's cross-path point, then subract 1.5s from my timings. That puts my D at the same moment as Kirk's, but my R is 0.7 seconds later. It's a physically plausible shape, but I'd like to know if the shape of Calvinia is already known from prior occultations.
 |
 |
 |
 |
 |
 |
 |
 |
The variations seen in the brighter tracking star might be due to some lingering cloud. The sky was quite full of cirrus just an hour or two before. This tracking star is the closest bright star to the target. |
PyOTE's solution is a duration of 3.4 seconds, which is almost 50% larger than the predicted duration. That's surprising. An unusually dark asteroid? Odd shape and I got the long axis? It does not look like a cloud extended the occultation, based on the tracking star's brightness. |
I had more time on Saturday 1/21/23, after a big meeting I'd had to prepare for. I did a re-analysis. This time, I narrowed my aperture to 5.3pix instead of the maximum 6.8 pixels. I also reduced noise by using "appsum" (i.e. no sky subtraction) in PyOTE and not using the tracking star as a reference. This gave much better timing accuracies, a higher S/N and also a rounder fit to the asteroid given Kirk's excellent data. This is the analysis that was submitted to IOTA on Jan 21.
 |
Raw data looks a sharper D and R, when using a smaller aperture. I worried that since the stars were not in focus, I must use a larger aperture, but 6.8 pix max was larger than needed and seemed to add more noise. |
 |
Re-analysis give a duration of 3.30 seconds, closer to Kirk's 2.8 sec, and tighter accuracies, and both timings moved closer to the direction of a rounder asteroid. |
Kirk Bender's Data
Observed from home, 2x setting on Watec. Got a 2.8 sec event, set gamma=0.8 which he saw as on screen as an improvement over gamma=1. Gain on max=41db.
PyMovie, all light curves. Target in yellow. |
PyMovie target light curve. |
No false positive |
Kirk's light curves are usually cleaner and better than mine, with the better camera - but this event, it was particularly striking. He had half the light per point as I did (2x vs. my 4x), yet still, his light curve is much sharper and better at the D and R sharp drops. |
PyMovie Screen capture. I note that his target star is much more visible than on mine, where I had to use a 222 frame finder stack in order to see my target well. I could see it on most frame-to-frame video while I set up, hence I chose 4x as my setting, but I was dismayed to see how dim it was on the actual playback and analysis. |
Kirk's driveway observing site. Nicely shows the tail of Cetus above the telescope, which is where the target was located. |
Reason for Kirk's much higher quality data? Once in a while, my light curve data is better, but rarely. Kirk's is uaully a little better. But this time it was dramatically so. Given past experience, I don't think simply a better camera could account for this.
* Did he have sharper images and better seeing? I didn't have the impression my focus was all that sharp, but I used the maximum size aperture: 6.8 pixels, which should have covered all of any spread out light.
* Cloud cover for me and not him? I didn't see any city light reflecting off of clouds. However, the stars did not look as bright as I would expect for the dark site I was at. The star looked unusually faint for a nominal 12.2 star. Was there pervasive thin cloud? But yet, Kirk was only 10 miles away, would it be perfectly clear for him?
* Was it that my gamma=1 vs his gamma=0.8? Maybe, but I've not seen gamma help me in my few tests. Kirk may be seeing what Dave Herald finds, and that is that lower from gamma=1 can help with faint stars.
1/21/23 - Solved. For faint out of focus stars like mine, data quality is better if you use 'appsum' in PyOTE, not 'signal'. In other words, don't do sky subtraction if the sky is so bright that it is adding noise in quadrature. New analysis shows Kirk's data still better than mine, but mine was significantly improved with a slightly smaller aperture on target and using 'appsum' not 'signal'.
I note that both Kirk and I had the star significantly bigger than the supposed maximum duration of 2.3 seconds. Mine was 3.3 seconds, Kirks was 2.8 seconds. This seems pretty solid. This would argue that the asteroid is either much darker than usual - which would be pretty unusual - or that it has a flattened shape and we caught the star traversing the fat direction, so that the frontal area was still similar to a smaller but rounder asteroid, which is where the max duration is calculated from.
