Sunday, July 21, 2019

#194 - Saturday, July 6, 2019 - In the Atacama Desert

After a few more days in La Serena following the solar eclipse, we started a three-day road trip up the coast and then inland to Calama, and then on to San Pedro de Atacama.  We stayed overnight in Chañaral and Antofagasta on the way up.  The desert landscape was like from another world!  There were many places where there was absolutely nothing but rock and sand.  And the occasional guanaco.  We got out at a guanaco observing spot near the Paranal Observatory, and sure enough, we saw one!  And then it made a funny noise and ran off.  Besides the crunching of our feet on the gravel, it was absolutely, perfectly silent.  We couldn't even hear the wind in this little valley we were in.  I had never heard such wide-open silence before; it was totally eerie and incredible.  A Girl Scout badge I led with a middle school group when I was in college talked about different kinds of pollution, including sound pollution, and it talked about the quietest square inch of space on Earth.  When a car wasn't passing by, this had to be it!


We were basically on Mars.  But with more oxygen.

What made it especially crazy was the fact that we were not at all far from the ocean.  In some places, we could see barren desert one one side of the road and crashing waves on the other!

The highway we were on took us past the Paranal site of the European Southern Observatory, and while we hadn't been able to get a tour there when we tried, we hoped there might be a cancellation, and at least we could see them from a distance if not.  We were out of luck, but the gate guard let us use the restroom there and re-fill out waterbottles, which was very nice.  We took plenty of pictures from the parking lot.  Yay for my long lens!

Paranal Observatory, from a distance

Posing with the sign

The altitude at the parking lot was around 7800 feet!  

In Antofagasta, our hotel was right on the beach, so we enjoyed watching the waves ceaselessly roll in, ever-changing, but ever the same.  It was hypnotizing.  

The Pacific Ocean -- the color of the water was beautiful!

On the third day of the road trip, Thursday, we drove for three hours from the coast up to the 7400 feet of the town of Calama.  We stopped at a large grocery store there called Jumbo (like Chilean Walmart, but a little nicer), since it was easier to get supplies there than in San Pedro de Atacama.  However, due to the sudden change in altitude, we were all lethargic, sick, and crabby.  So we tried our best to hold ourselves together while we shopped quickly, and then drove the next hour to San Pedro de Atacama, and the Atacama Lodge.

The Atacama Lodge is an amateur astronomer's dream land, owned by French ex-pat Alain Maury, himself an astronomer who grew tired of the light pollution and the politics of France and sought better skies.  It grew to be part robotic scope operation, part astro-tourism, a few kilometers south of the touristy village of San Pedro de Atacama in northern Chile, lying in the shadow of the volcanic Andes Mountains.  He hand-built a sunflower field of telescopes, with which he hosts nightly tours in three languages to showcase the wonders of the Southern skies.  


Less of a hotel and more of a nice set of adobe cabins, the accommodations were charming and fun.


My room

John & Beth's room

Gorgeous mosaic tile in the bathroom

Kitchen & dining area, complete with cookware and dishes.  And, very importantly -- a coffee maker!

Actual thatch roof!

Bonus, it had a heater, which none of our previous hotels had, which was a godsend.  The forecast showed that it would dip into the 30s every night!

Once Alain himself finally returned from a grocery run to Calama, he quickly got me set up on the Sky-Watcher AZEQ6 with a Takahashi FSQ-85ED before he had to zip off for that night's astro-tours.  The mount was located in a giant motor-driven rolloff shed that contained several robotic scopes owned by people from around the world, including one owned by none other than Guylain Rochon, the man behind the BackyardNikon and BackyardEOS camera-control software that I use.  Since the shed automatically opened and closed at sundown and sunup, I'd be able to leave everything connected, and it would be kept safe from the heat and the dust during the day.


Alain ran back to his house and got the key for the little half-door, but couldn't get it to unlock.  After much effort, including attempting to remove the lock from the door (he just jumped over the half-wall to get to the inside), we finally just set up a chair on the inside and outside to use to help climb over the wall!  I got pretty good at it by the end of the night in the dark.

The Takahashi didn't have a guide scope attached, but Alain said it wouldn't need it, and I hoped he was right.  I had prepped the software on my computer for plugging the hand controller in via serial-to-USB cable and running it using the SynScan driver and app, but he didn't have that cable -- instead, the AZEQ6 had a direct USB cable, like the Celestron CGX-L mount, which I had no idea it had.  This required EQMOD to run.  Before it got properly dark, I went back to the wifi in the lodge to install EQMOD, but I couldn't get the driver to show up in ASCOM when I tried to connect to it in Cartes du Ciel.  So instead, since I wasn't guiding, I just used the hand controller and didn't bother with trying to control it from my computer.  I just hoped the gotos would be good enough to not need a finderscope, which it also didn't have.  I got my ZWO ASI1600MM Pro and Starlight Xpress 2-inch filter wheel attached without issue.  Alain had a Takahashi-to-T-thread adapter handy (although I had brought mine just in case). 

My ZWO ASI1600MM Pro camera and Starlight Xpress filter wheel attached to Alain Maury's Takahashi FSQ-85ED on a Sky-Watcher AZEQ6 mount.

Sunset was gorgeous, especially the way the Andes changed colors.  I have a timelapse video that I'll put together (eventually), which I'm really excited about.


Once it got darker, I needed to get the camera focused.  I had it slew to a bright star, but since I wasn't sure how well polar aligned it was at the moment, that star did not end up in the field of view of the camera.  I cranked up the screen stretch on the histogram in SharpCap, but still couldn't even see any giant blobs.  So instead, I slewed it myself over toward some patch of the Milky Way, where I knew there would be stars in the field no matter where I pointed.  After over-stretching the histogram, I saw some giant blobs, and was able to get the scope roughly focused for alignment.  I would like to take a moment to point out something amazing -- this is a one-second exposure.  There were so many stars!  It was so dark that many more stars than usual could stand out.


Wowee!!
Next, I ran the polar alignment process in SharpCap, which it was capable of doing in the southern hemisphere the same way it does in the northern -- by way of plate solving.  


It was during this process than I ran into my next stumbling block -- Alain had showed me where the RA and dec clutch knobs were, but I couldn't remember where the RA one was.  So I tried various things for about 15 minutes before I finally figured it out!  But I finally got the darn thing polar aligned.  I also balanced it before polar aligning, although I wasn't able to balance it in dec because I ran out of dovetail to move it any more forward to compensate for my camera gear on the back.

Next was alignment, which I did with the hand controller in a very similar process to Celestron.  The stars landed in the FOV for every star, which was very exciting and helpful.  Finally, after focusing on a bright star with the Bahtinov mask I'd brought with me, it was time to line up my first target.  I chose something easy -- compact and relatively bright -- for my first target: Centaurus A.  It's a large, oddly-shaped galaxy, somewhere between 10-16 million lightyears away, also sometimes called the Hamburger Galaxy. Viewing it visually makes that moniker choice obvious -- it was a bright upper and lower hemisphere, split by a thick dark band of dust, making it look very much like an enormous hamburger in the sky.  You can see it from the southern United States -- from the 29 degrees north latitude of the Texas Star Party, it crests at 12 degrees off the horizon or so, just high enough to let it peak for an hour or two above the hills.  But here at 22 degrees south, it was quite high in the sky, with much less atmosphere between me and it.  Thanks to the dark skies, it was very easy to see with 1-second exposures in SharpCap, so I centered it by hand, switched over to SequenceGenerator Pro, keyed up the sequence I wanted, and let it run.  The plan I had developed several months ago for which targets I was going to image when called for starting with RGB that night before changing targets for the second half of the night -- with RGB, I'd have a complete dataset, in case something broke later in the week and I didn't get to finish.  Having a luminance channel is great, but not required.  

I took a few test frames first to see how long it would track for, and I got as far as 5 minutes with just the tracking, and my stars were still mostly round!  I set the exposure time to 3 minutes so I could get more frames, and because that was long enough for the ZWO in these dark skies with this brighter target.  

Single 180s red channel frame, screen-stretched & zoomed in

The RGB frames looked a little out of focus, so partway through the night, I slewed over to a nearby bright star and checked the focus with a Bahtinov mask.  I tweaked it slightly, but it was very close.  I cinched down the focus lock a little tighter, slewed back to Centaurus A, and carried on.

Single 180s blue channel frame

While the ZWO was running, I set up my Sky-Watcher Star Adventurer with my Nikon D5300 and 70-300mm lens attached again, which I found some space for in the shed as well. Polar aligning this one was going to be harder -- I had no SharpCap to help me.  (I am working on putting together a rig for one of my guidescopes that will attach to the second camera screw on the Star Adventurer, which I could use both for polar alignment and for auto-guiding.  Yes, the Star Adventurer has a ST-4 port!)  I used my cell phone to get it pointed due south and get the elevation close to correct, but I couldn't see the Octans trapezoid in the polar scope still.  So I pointed it to the eastern horizon and then to zenith to do some drift alignment, which helped a bit.  I pulled the zoom on the lens back to something around 135mm, focused it, and then turned it to the glorious Eta Carinae Nebula.  I was able to get about 90s exposure times with minimal streaking, so I set it there and let it run.

Later in the night, I had a brainwave -- the Takahashi clamshell ring had a 1/4-20 tripod screw on the top!  So I took one of my ball heads and attached it to the top, and grabbed my Nikon D3100 and my new 35mm f/1.8 Nikon lens and got everything attached.  I aimed it up somewhere in the middle of the Milky Way (yes, up, overhead!!), vaguely toward M8 & M20, and took a 30-second test exposure at f/2.  I went to check on my D5300, and caught a quick glimpse of the D3100 image once it had completed and flashed onto the screen, and I froze.  I stood there for a moment, dumbfounded.  Did I just see what I think I saw?  I fumbled for the Review button, and it re-appeared on the screen. Holy McShitNuggets (were the actual words that came out of my mouth), it was incredible!

30s exposure pointing up toward the Milky Way with my Nikon D3100 and 35mm f/1.8 lens at f/2.

So I just left the camera pointed there, set it to 60s exposures at ISO-800 and f/2, and let it roll!

After all this work, everything was clicking away happily, so I went back to the lodge to warm up, take a break, and have a snack.  After the break, John, Beth and I wandered over to a lone 24-inch Dob beside another one of the cabins, which was being used by part of a large French group that was also visiting the Atacama Lodge.  Between people, we snuck in to observe the massive globular cluster 47 Tucanae.  Located about 13,000 lightyears away, it has about 10,000 stars packed into a ball about 120 lightyears wide.  It's large and bright enough to be easily seen naked eye just off the edge of the Small Magellanic Cloud, and is second only to the Omega Centauri cluster.  47 Tuc may even have a black hole at its center.  It spans an area of the sky that is wider than the full Moon by almost twice!  It was so spectacular in the large telescope that it prompted me to shout "Holy mackerel!" (yes that is what I actually said), which garnered some chuckles from the French.  

We walked up to the main field to nab a scope up there to use for the night, but the French had commandeered all of them.  So I went back over to my cameras to check on things.  Eta Carinae had gotten too low to continue imaging, and it was time to change targets on the Takahashi as well, so I took the opportunity to swap lenses between the Nikon D3100 and D5300 so I could put the longer lens on the mount that was tracking more precisely.  However, this proved to be too heavy for the Sky-Watcher AZEQ6 mount, for which I already had the counterweight at the bottom of the rail, and gravity overcame the RA clutch knob and the scope slipped over!  I caught it, and it was not like it was totally loose or anything, but it did mean that I had to start over with alignment.  I tried to tighten down the RA axis some more, but I couldn't get it any tighter.  Alain had said he'd look for a second counterweight in the daylight.  So I took the camera off, re-aligned (gotos weren't as good this time, however), and pointed it to my second target for the night, a galaxy in the southern constellation Pavo called NGC 6744.  

I started with the L channel this time, but it looked out of focus.  So I checked focus again with a Bahtinov mask, but it was still perfect.  I thought that maybe it was evenly-sided periodic error or something.

180s luminance image of NGC 6744, zoomed in

I pressed on anyway, since I wasn't sure what else to do.  Then I went over to my Nikon D5300 and set it up on the Small Magellanic Cloud with the 35mm lens.

Around 3:30 AM, I was petering out.  The long day and high altitude were making me extra-tired.  So I shut down the power to everything like Alain had asked me to, and got into bed around 4:15 AM.

It was a great night!  I got my imaging rigs rolling, the sky was incredible, and the Milky Way was simply mind-blowing.  We couldn't quite see any color, but the transparency might not have been the clearest that night due to the earlier clouds.  There were also some clouds still down near the horizon.  But the Large and Small Magellanic Clouds that appeared later in the night were easy to spot, as well as a large number of deep sky objects with averted vision.  I couldn't wait for the rest of the week!  We had five full nights under these skies!

[ Update July 21, 2019 ] 

I processed this image last weekend, but now that I finally got this log entry written, here is the Eta Carinae Nebula!

Date: 6 July 2019
Location: Atacama Lodge, San Pedro de Atacama, Chile
Object: Eta Carinae Nebula
Attempt: 1
Camera: Nikon D5300
Telescope: Nikon 70-300mm lens @ 90mm, f/4.5
Accessories: N/A
Mount: Sky-Watcher Star Adventurer
Subframes: 54x90s (1h21m)
Gain/ISO: ISO-1600
Acquisition method: Intervalometer
Stacking program: PixInsight 1.8.6
Post-Processing program: PixInsight 1.8.6
Darks: 0
Biases: 0
Flats: 0
Temperature: 33-36F

The stars came out a little bloated, and it turns out I was closer to 90mm than the 135mm-ish I thought I set it to, but isn't it a gorgeous region??  

Featured here is the Eta Carinae Nebula, which is a massive stellar nursery located in the constellation Carina, or "keel of the ship."  It's about 8,500 lightyears from us, and is 460 lightyears across!  From our perspective here on Earth, it's four times larger and is brighter than the famous Orion Nebula!  Unfortunately, however, you can only see it from southern skies.

It has a multitude of interesting stellar objects contained within.  One such object is the star Eta Carinae itself, which is an extremely luminous hypergiant star, weighing in at 100-150 times the mass of our Sun, and outputting four million times more light.  Due to its large mass, it is likely to go supernova in the near future ("near" in astronomical terms, of course). 

Immediately surrounding the star Eta Carinae is the Homunculus Nebula (probably the coolest nebula name ever, meaning "Little Man" in Latin, which sounds much less cool), which is the result of a huge outburst from Eta Carinae seen on Earth in 1841, which for a short time made it one of the brightest stars in the sky.  It's a very oddly-shaped nebula, and there are some very cool pictures out there of it!  

Also pictured here are the Southern Pleiades to the left, and the tighter open cluster NGC 3532 (also called the Pincushion Cluster, Football Cluster, or Wishing Well Cluster) to the upper right, as well as several other clusters.  

Here's how I processed it in PixInsight:

- Since I don't have any 90s dark frames at those temperatures, I skipped right to SubframeSelector
- Image scale: 8.97 arcsec/px
- Gain: 0.115 e/ADU
- Highest scoring frame: DSC_0050 (88.648)
- Debayered
- Registered with StarAlignment, with DSC_0050 as reference
- Stacked with ImageIntegration, using Linear Fit Clipping for rejection
- Cropped with DynamicCrop
- Applied DynamicBackgroundExtraction
- Color corrected with PhotometricColorCalibration
- Tried Deconvolution, but didn't like how it made the stars look
- Denoised with MultiscaleLinearTransform
- Stretched with HistogramTransformation
- Applied HDRMultiscaleTransform, 7 iterations, with lum mask
- Touched up with CurvesTransformation and HistogramTransformation

I'll probably go back and re-process this one and try to get the stars to not dominate the image as much.  We just discussed this on The Astro Imaging Channel last week, so I'll have to go back and re-watch it.

[ More update July 21, 2019 ] 

After I posted some of the screenshots of the Centaurus A galaxy, I realized the data didn't look as bad as I thought.  So I spent the later part of the afternoon processing that data.

I realized when I was filling in my documentation text file that somehow the gain got set to only 10 for this dataset, for which I have no dark frames.  I tried processing with gain 0 frames, but the calibrated light frames came out pretty much nuked of light.  So I went ahead and processed without calibration frames, but they turned out not to be that necessary, since I had my sensor cooled to -30C.  I didn't take any luminance data, although in a future re-do of this dataset, I might crop and re-scale this color data and add it to the luminance data I took of Centaurus A at the Texas Star Party.  It may or may not help, since Centaurus A was much lower in the sky -- and thus in the thicker, nastier part of the atmosphere -- but it might be worth a shot.

Anyway, here it is!

Date: 6 July 2019
Location: Atacama Lodge, San Pedro de Atacama, Chile
Object: Centaurus A
Attempt: 2
Camera: ZWO ASI1600MM Pro
Telescope: Takahashi FSQ-85ED
Accessories: Starlight Xpress filter wheel, Astrononik LRGB Type 2c 2-inch filters
Mount: Sky-Watcher AZEQ6 (Alain Murray's)
Guide scope: N/A
Guide camera: N/A
Subframes: R: 28x180s
   G: 26x180s
   B: 10x180s
   Total: 64x180s (3h12m)
Gain/ISO: 10 (not sure how that happened)
Acquisition method: SequenceGenerator Pro
Stacking program: PixInsight 1.8.6
Post-Processing program: PixInsight 1.8.6
Darks: 0
Biases: 0
Flats: 0
Temperature: -30C (chip)

This might be only my 'official' second attempt, but I had tried to image it before but ran into other issues that led me to not keep that data.  So I'm happy to finally have an image of this cool galaxy!  It's not only cool to look at in its weirdness, but its weirdness has made it an interesting target of study.  Centaurus A is somewhere between 10-16 million lightyears away -- no one is quite sure.  It doesn't fit cleanly into any established galaxy category, and its odd shape is probably a result of a galaxy merger.  At the center of that galaxy is a 56-million-solar-mass black hole that is blasting out jets that move at half the speed of light and emit brightly in x-ray and radio.  They extend out to a million lightyears!  Very cool galaxy to look at.  It's also on the top-5 list of the brightest galaxies in the sky as seen from Earth, so it makes a good visual observing target.

[ Update August 8, 2019 ] 

Still plowing through Chile datasets -- here's the Magellanic Cloud!  It turned out to be just a hair out of focus, so the quality isn't that great, but here it is anyway.

Date: 6 July 2019
Location: Atacama Lodge, San Pedro de Atacama, Chile
Object: Small Magellanic Cloud
Attempt: 1
Camera: Nikon D5300
Telescope: Nikon 35mm f/1.8G @ f/2
Accessories: N/A
Mount: Sky-Watcher Star Adventurer
Guide scope: N/A
Guide camera: N/A
Subframes: 28x90s (42m)
Gain/ISO: ISO-1600
Acquisition method: Intervalometer
Stacking program: PixInsight 1.8.6
Post-Processing program: PixInsight 1.8.6
Darks: 0
Biases: 20 (28F)
Flats: 0
Temperature: 32-35F

I did try to apply some deconvolution to see what sharpening could be done on the SMC, and with a small number of iterations, it helped a tiny bit.  I did try MaskedStretch for stretching the image, but it blew out the small pink and blue halos around the stars, which I didn't like.  So I re-stretched it myself with HistogramTransformation instead.

Here's the whole process:
- No darks, calibrated lights with pre-made superbias using ImageCalibration
- SubframeSelector:
- Scale: 23.07 arcsec/px
- Gain: 0.115 e/ADU
- Highest-scoring frame: DSC_0162 (85.313)
- Debayered
- Registered with StarAlignment
- Stacked with ImageIntegration
- Combination: Average
- Normalization: Additive
- Pixel rejection: Linear Fit Clipping
- Cropped with DynamicCrop
- Applied DynamicBackgroundExtraction
- Denoised with MultiscaleLinearTransform
- Color corrected with PhotometricColorCalibration
- Applied Deconvolution using DynamicPSF, range_mask-star_mask, 10 iterations
- Tried MaskedStretch, but stars got pink and blue halos, so I stretched it myself
with HistogramTransformation
- Tweaked with CurvesTransformation
- Applied HDRMultiscaleTransform, 9 iterations

[ Update August 15, 2019 ] 

Still chugging through Chile data!  Being in the middle of a move makes finding time and energy to do this difficult!  But I finally had some chill time to work on it.  

I processed the stack of the Milky Way I took with the 35mm lens piggybacked on the Sky-Watcher AZEQ6 mount -- the "Holy McShitNuggets" dataset -- and the result was truly excellent!  There was also almost zero drift across the whole time of the stack, which was awesome and gave me nice round stars.

Date: 6 July 2019
Location: Atacama Lodge, San Pedro de Atacama, Chile
Object: Milky Way
Attempt: 9
Camera: Nikon D3100
Telescope: Nikon 35mm f/1.8G @ f/2
Accessories: N/A
Mount: Piggyback on Takahashi FSQ85ED, on Sky-Watcher AZEQ6
Guide scope: N/A
Guide camera: N/A
Subframes: 39x60s (39m total)
Gain/ISO: ISO-800
Acquisition method: Intervalometer
Stacking program: PixInsight 1.8.6
Post-Processing program: PixInsight 1.8.6
Darks: 0
Biases: 0
Flats: 0
Temperature: 31-35F

I just can't get enough of that Milky Way!!  There are several interesting things here.
- The super bright thing on the right is Jupiter
- In the lower right is the Rho Ophiuchi complex, which I've recently shared closer-up images of. The brighter yellow-orange star is Antares.
- The dark upside-down-U-shaped nebula to the left of Jupiter is the Pipe Bowl Nebula, a giant molecular cloud
- The two fuzzy red regions in the center of the image are two emission nebulae - NGC 6357 on top, and the Cat's Paw Nebula on bottom
- Up above NGC 6357 is open cluster M7, Ptolemy's Cluster
- In the lower left of the image, the other red nebula region is emission nebula IC 4628, surrounded by stars that are part of the Trumpler 24 open cluster

...and much more, of course!

This one was a pretty quick job.  Being one-shot color and not having issues or light pollution helped!
Here's the PixInsight processing:
- No darks or biases b/c I don't have a library for my D3100
- SubframeSelector:
- Scale: 25.9 arcsec/px
- Gain: 0.115 e/ADU (not actually measured, just borrowed from D5300)
- Debayered
- Registered with StarAlignment, using highest-scoring frame was reference
- Stacked with ImageIntegration
- Combination: Average
- Normalization: Additive
- Pixel rejection: Linear fit clipping
- Tried skipping DynamicBackgroundExtraction and did PhotometricColorCalibration instead, since there's pretty much no light pollution and most of the image is Milky Way anyway
- Denoised with MultiscaleLinearTransform, with luminance mask
- Stretched with MaskedStretch
- Tried with target background 0.15, interesting result, but not a huge fan
- Tried again with target background 0.10
- Tried again with 0.05, but too low, so did 0.08 - let me try some curves on that
- Adjusted with CurvesTransformation
- Applied the DarkStructureEnhance script

I wanted to have nice contrast between the bright and dark areas, and show just the insane number of stars in the bulge.  I also wanted the dark nebulae to stand out nicely.  But I didn't want it to look too hard or electronic either -- kind of softer, like how it might appear to our eyes if they were more sensitive.  As with many of my images, I'll probably mess around with it more in the future.  But I am quite pleased with this one!





2 comments:

  1. Nice image Molly! I just saw this on Astrobin and found your blog. Having fun reading it. Nice eclipse photos too!

    ReplyDelete