A Dog, A Dragonfly, An Experiment, and Variables

Lets start with the dog; a Disney dog in this case. Specifically Pluto. If it can't be a planet, I guess it can be a dog. One of my observing goals has been to “see” all of the planets in our solar system. The last one left would be, of course, Pluto. I know, I know; Pluto isn't a planet any more. But it was when I started observing, so Pluto it is. I can tell you up front, it's spectacularly unimpressive.

Pluto, somewhere in there

Above is an image showing Pluto. It's that tiny dot of light inside the white box. See what I mean? Spectacularly unimpressive. The hunt to find it took quite a long time. But it was fun. Been there, done that...Moving on....

The remaining images in this entry are experiments. The next image of NGC457, also know as the Dragonfly Cluster (or the Owl Cluster, or the E. T. Cluster or the Kachina Doll Cluster, among others; take your pick), was an attempt to overcome problems with light pollution by taking a large number of exposures. The expectation was that it would help to increase contrast, but wouldn't be a substitute for a dark sky site. Pretty much, that seems to be the case. It also became an unintentional segue into working on variable stars and problems with at least one of those. More on that later.

NGC457

NGC457 is an open cluster in the constellation of Cassiopeia. It's about 7900 light years away and quite nice, I think. The two brightest stars are often imagined as eyes and the rest of the stars (about 150 or so in the cluster) as “something”; E. T., a dragonfly, an Owl, or whatever. Technically, the image is quite a challenge; most of the stars in the cluster are quite dim, magnitude 12 to 15. The two bright stars are magnitudes 5 and and 7. Why is this a challenge? (Thank you for asking!) Well, the difference of one magnitude is a difference of 2.51 in terms of brightness. That is to say, if there is a difference of 1 magnitude between two stars, one of them is 2.51 times brighter than the other. So, if the dimmest star I captured is, say 12^th magnitude, the difference between the brightest star (5^th magnitude) and dimmest start (12^th magnitude) is 7 magnitudes. That means the brightest star is 2.51x2.51x2.51x2.51x2.51x2.51x2.51= 627 times brighter than the dimmest star. That's a lot! That difference is referred to as dynamic range. If you consider that the sky is actually the darkest element of the image, the dynamic range is actually much greater than just the difference in magnitudes of the stars. To capture the dimmer stars, longer exposure times are required, which causes the brighter stars to be overexposed or saturated. When that happens with CCD cameras, the overexposed pixels “bleed over” into adjacent pixels. Think of filling a glass with water, which gets overfilled and spills over. The name given to this is “blooming”, and cameras like mine have circuitry called anti-blooming circuitry to help with this. As it turns out, I needed to retouch the image to reduce the blooming that occurred around the bright stars. The original image shows the blooming.

The next experiment involves a different camera, one that's called a “one shot color” camera. It's the one I use for lunar imaging, and is essentially a color webcam. This one has the provision for internal stacking (adding successive frames to make one still frame) to make a long exposure. The target this time was another bright open cluster, M11, also know as the Wild Duck Cluster.

M11

One of the problems with using this camera is the inability to guide the telescope automatically. All telescopes that track the sky do so with some time of mechanical gearing. Because gears can't be made perfect, they cause tracking errors that show up in the image as stars that are not round, but elongated in one or two directions. To correct for that as much as possible, some type of guiding is used. The easiest type is another camera that sends the correcting signals automatically. That's the method I usually use. So, images taken with this camera will always have these guiding errors. That's usually not too bad, because the exposures are usually very short, so little movement is involved. However, it will show up in as little as 10 seconds with the image scale (or apparent magnification) used in this image. What I was able to do was get 10 images, 9 of which were usable, each 10 seconds long to make the single image you see above. Bottom line, I might be able to use this method for bright star clusters and get a color image quicker, but not with as much fine detail, as using the “bigger” camera.

Finally, one of the goals of the Starlight Observatory is to be able to accurately record the brightness of the class of stars called variable stars. As the name says, these stars change their brightness over time, and observations of that variation helps to determine what the cause of the variation is. One way to measure how bright a star is would be to measure the number of electrons in the pixels that are lit up by the variable star, and then do the same thing for a couple of other stars in the same image that have known and documented magnitudes. The process would work like this: 1) take an image showing the variable (V) star, and the two other known stars, called the check (K) and comparison (c) star. 2) the camera software converts the number of electrons into a number between 0 and 65535. 3) Then, subtract the “software”, ie, 0 to 65535 value, of the Check star from the Comparison star. 4) Subtract the “software” value of the Comparison star from the Variable star. Mathematically, K-C and V-C. To find the value of V, I just need to add the known value of C to the value V-C and I've got the value for V. K-C serves as a check to be sure all is ok. Example: Known values for K and C are K=13.956, C=12.743, K-C= 1.213. Observed Value for V-C (ie, the value derived from my image) = 1.791. The derived value for K-C = 1.052. For my data, I get the value, or magnitude of V as 1.791+12.743= 14.534. How close is that to the “real” value? One measure is the K-C value. If it matched exactly, I could be pretty sure. However, there is the difference, 1.213 vs 1.052 so confidence is not perfect. There are explanations for the difference (variation in the cloud cover, camera movement, inexact flatfielding, which is really the most probable cause). But for now, just being sure I'm in the right ball field is good enough. Well, the thing about most variable stars is that they vary over fairly long time periods; weeks, months, or years. There are, however, a few that vary over much shorter time spans, like a few days. Algol is one such star. So, on the night of October 16^th, after checking the Sky and Telescope Algol predictor app, I head out to the observatory to record Alogl's variations. From my point of view, Algol (ra's Al-ghul, in Arabic) lived up to it's namesake as the Demon (ghoul) Star. Either the S & T app was incorrect, or my camera just couldn't record the variation, but I got 2 hours of images similar to the one below.

Algol, The Demon Star, Appropriate for October

The brightest star is Algol. And by my calculations, it never changed brightness significantly. It should have decreased in brightness by over a magnitude ( more than 2.5 times). It could have been my camera; the shortest exposure it can take is 0.1 seconds. Even with that short of an exposure, Algol is still out of the linear response area of the camera (a bad thing for recording variable stars) and very near saturation of the image (meaning overexposed). It was this result that lead me to try using the webcam for star exposures. It can make a much shorter exposure to prevent saturation. Problem is that I would need to do more rigorous testing of the camera to determine it's linear region. I'm not sure I'm ready for that.

Well, that catches me up on all the images for today. As we go into the winter months, we go from “globular season” to “open cluster season” to “galaxy season” and back to “globular season next year. Interesting, and some not so interesting nebulae thrown in for good measure along the way. We'll see what the sky holds for next time.

“Previously Undocumented Feature” found on the Moon

OR, how to make a completely serendipitous find look like the culmination of a life's work

The months of May, June and most of July had been unfavorable for astronomy at the Starlight Observatory. Finally, in an act of frustration and defiance, I awoke about 5:30 AM on July 24, 2016 to go out to the observatory to see something. Since it was last quarter moon, I knew I would probably end up looking at the moon, but decided to start elsewhere. I really don't remember much of what I viewed, but I did end up on the moon. I usually like to “run the terminator”, meaning I like to view what's on the terminator, since that's where the shadows will delineate the high and low areas the best. This time, however, I noticed something I had never seen before, or at least I didn't remember seeing before. Between Montes Spitzbergen and Mons Pico was a dark area that I didn't recognize. So I decided to take an image of the area. For those of you who like to read the end of the book, and then fill in the details, it would appear that I have found a “previously undocumented feature” on the lunar surface. Is it scientifically important? I think the answer is, I don't know, but probably not. It's just “interesting”. But, for me, exciting. What follows is what I've managed to document to date.

After closing the observatory and coming inside. I processed the image to see exactly what I had. After processing, I decided I needed to find a lunar map to decide if I was really seeing something I hadn't seen before or not. The lunar map I used was the book by the Soviet lunar cartographer Antonin Rukl. It serves as a pretty good reference book of the features on the moon. The area of interest was to be found on plates 11 and 12 of his book. Plate 12 contains most of the area of interest. Shown below are plates 11 and 12 in a composite image made using Paint Shop Pro 7.

I couldn't see anything that looked like my image.

Shown below is my image and plate 12 from Rukl's book, roughly to the same scale. You can match up Montes Spitzbergen and the crater Kirch in both images when viewed this way.

I can see some shading in Rukl's map, but I think that is because of the difference in lunar soil reflectivity. Looking closely, his shading does not exactly match the contours of the area I have imaged. And I think I know why. The answer most likely can be found in the image below.

Time to jump ahead in the story. My image was reviewed by 2 members of the Association of Lunar and Planetary Observers (ALPO, for short... not a dog food, by the way). It was these members that informed me that this is a “previously undocumented feature”. However, the important aspect at this point is the telling statement made by one of the members that what you see above is “the way most people see the moon when looking for features.” If you are not familiar with lunar observing, the thing to notice is that the sun is too the right in this image. The orientation of the image is that same as if you were looking at the moon with just your own eyes. That is to say, north is roughly up,and east is to the right. The terminator has “night” to the left and “day” to the right, so this is the morning terminator. This is the way you would see the moon if you went out after supper (I know, “supper” is a southern expression. But I'm in Georgia. What do you expect?) to look at the moon. In fact, this image was taken about 8 PM on October 8, 2016. Now that you know exactly where to look, you can see most of the valley as a difference in “color” or shade of gray (this is a black and white image). Now compare this image (sun to the east), with the original image, with the sun to the west.

Now the valley is very obvious. My crude measurements show it to be about 95 miles long, along its long (north-south) axis. It appears to be about 1/5 as wide, so it's maybe 20 to 25 miles wide, with a steeper slope on the eastern, or right, side. It appears to be about 300 feet deep. So the reason I suspect it hasn't been “seen” before is that if wasn't observed with the sun illuminating it (the valley) from the west and just along the terminator. In other words, I was simply at the right place at the right time.

What about a confirmation of this feature? That's the process I'm in now. The place I'm looking is at data, specifically a map, made by the Lunar Orbiting Laser Altimeter; LOLA, for short. I don't know if I can get the resolution of the LOLA map fine enough to place the valley on the LOLA map in the correct place to specifically confirm the finding, but I strongly suspect it will. The LOLA map does show a depression is the general area and nothing else seems to fit.

As an aside, look again at the image comparing my image to Rukl's plate 12. You might notice my image shows several craters that are not on Rukl's map. The two explanations that come to mind are 1) they weren't there when Rukl made his map, and there is recent evidence that there is quite a few craters being discovered since the 1970s, (see http://www.space.com/34372-new-moon-craters-appearing-faster-than-thought.html) or 2) they were there and Rukl decided to not include them for some reason. I have no idea which is true.

So. there you have it, I guess. I wonder if anything will ever become of it. For my part, it's been a fun adventure for the last 2 months.

Last Day of Summer, and It's Globular Season Part 1

At least, probably Part 1 (and published later than 9/21....busy, busy, busy!)

Globular Season refers to the vast number of globular clusters visible in the sky. A globular cluster is a large, spherical cluster of stars. They are typically found orbiting the center of a galaxy, and, apparently, formed about the same time as the galaxy. They are typically the oldest stars in (or around) a galaxy.

 Above is a screen shot from the program Stellarium showing the area around the constellation of Sagittarius. The center of our galaxy is very near this constellation. The circles with an + in the middle show globular clusters. Obviously, there are quite a few.

This is the same image, but the box shows encloses the constellation of Sagittarius. There are more than 20 globular clusters within the box. The clusters shown below are all within the box and are all members of the Messier catalog. All images are in black and white. I took the images in color, but there was so little color shown that I think they “show” better in black and white.

M22

M22 was the first globular cluster to be discovered in 1665. The discoverer was Abraham Ihle. It is about 10,000 light years away from us, making it one of the nearer globular clusters. Visually, it's about 17 arc-minutes in diameter. If all the associated stars are included, it's about 32 arc-minutes in diameter, making it larger than the full moon. It it the brightest globular cluster. M22 is notable for two other discoveries: 1) it contains a weak planetary nebula cataloged as IRAS 18333-2357. This was the second planetary nebula discovered in a globular cluster and one of only four known planetary nebula in Milky Way globular clusters. 2) Recent Hubble Space Telescope investigations have discovered “a considerable number” of planet-sized objects that appear to float through the cluster.

M28

M28, which is about 18,000 to 19,000 light years away, is smaller than nearby M22. It was discovered by Charles Messier in1764. M28 is the second globular cluster where a millisecond pulsar has been discovered in 1987. The pulsar spins around its axis once every 11 milliseconds.

M69

M69, as well as M70 below, were both discovered by Charles Messier on the same night, August 31, 1780.

M70