Two weeks ago I went down to Kitt Peak National Observatory in Arizona with my advisor to do some observing. Physics as a discipline isn’t really prone to sending students on trips (other than to conferences or to work with collaborators on a multi-university project), and astronomy only for observing trips like the one I just did, and consequently it was only the second time I’d gotten to travel for school/work/whatever grad school is. I did more theoretical stuff when I was an undergrad and I’ve been using archival data since I began in observational astronomy so going anywhere just hadn’t happened yet. This semester though there has been a lot of observing activity in our group, so even though it isn’t specifically connected to my thesis work, we’ve needed people to help out. In this case for a project involving star formation rates within galaxy clusters.

The local environment within clusters of galaxies (presumably some aspect of its density at the point in question) has an effect on the star formation rate within the galaxies. Since there haven’t been a ton of observations of the lower rate, fainter, galaxies in clusters the idea was to come up with a method of imaging them so we could gain some insight into how many stars are forming there and why. We chose a few known galaxy clusters that are at a specific distance such that the primary emission wavelength is redshifted to be just within a narrow-band filter on the MOSAIC imager of the telescope we used (the Mayall 4-meter). Then, by comparing the difference between the narrow and wide band filters, we could pick the galaxies that were at this distance out from the background, and figure out their rate of star formation.

So that’s basically the science justification for what we were up to. Like I said, it isn’t specifically what I’ve been working on myself (though I do study galaxy clusters) so when I say “we” I really meant “the PI and my advisor” (at least in the last paragraph).

So back to focusing on me.

I had to leave from Boston at like 9am, so even though I knew I’d be up all night observing, I had to get up at like 6 am and spend most of the day in seats that weren’t designed for 6’1”, leggy brunettes like myself. We basically had to rush to get there in a timely manner, the observatory is about an hour outside Tucson, and my flight didn’t get in there until about 90 minutes before we had to get down to business

For scale, think of each letter as being the height of an average human. It would be incorrect, but think about it anyway.

The telescope we were using for this project was the Mayall 4-meter at Kitt Peak. It’s a beast of an observatory— the telescope, the dome it’s in, and the structure itself, are all enormous. At least, they feel that way considering that it’s about 15 stories high and there are usually only 3 people working in there. It was built in the early 70s when the thinking was that getting any height above the ground was an improvement in seeing (the clarity of images), and so even though it was already on top of a mountain, they deliberately made it wicked tall. And to accommodate the the size of the telescope’s focal length (as you can see above) the inside of the dome, and therefore the footprint of the building had to be large as well. As it turned out, it really didn’t matter how tall it was, and in fact, giving it this massive structure caused wind coming up the side to shake the telescope a bit, which is obviously bad. Nonetheless, it’s a great instrument, and powerful enough to accurately resolve some faint galaxies.

The first night was really painful. After having been up for about 30 hours in a row, (especially when a lot of that time consists of traveling-while-tall) everything in my body hurt, and for about an hour around 4 am my consciousness was flipping on and off like a light switch. So good job there, self. And coming out at about 6:30 am onto a freezing, surprisingly windy mountaintop was fairly painful as well. Fortunately, it was worth it since we got so much data that night. My advisor, despite working around the clock and having a couple kids, has that Margret Thatcher superpower where he only needs a few hours of sleep every day (his approval numbers are better among Scottish miners too). So luckily for us, he breezed through the all-nighter. If it had just been me, I probably would have collapsed on the controls around 4:30 and caused some kind of comical mishap.

As I said in the little science part, in addition to two common filters, we were taking images in two narrow-band filters which required us to have the shutter open for 10 or 20 minutes at a time. Once you’ve gotten set up and conditions are basically constant, and you are doing a 5-image pattern of 20 minute exposures, it definitely gives you some time to think. So I thought about how to write a python script sorting data based on header information in the files I’m using, and since I was so tired and achy, it took about 5 times as long as it should have, and involved deep thoughts about how quotation marks in certain contexts cause syntax errors.

One of the cool things though about the ultra-long exposures is that they are actually long enough to detect asteroids. You see all the usual stars and galaxies, and since the exposure is so long, a higher than average number of cosmic rays, but if you look closely, there are numerous little streaks, all going in the same direction. The Earth’s motion relative to the asteroids is such that over the course of 20 min we have moved far enough relative to them that they’ve steadily moved through the background. On the third night we accidentally set the telescope to track on an asteroid, and got the reverse effect. You pick a bright guide star on each exposure and it helps the instrument to follow the field through the Earth’s rotation. It was a 13th magnitude asteroid, and the operator understandably mistook it for a star, so after 10 minutes when we saw the image, everything was streaked out, with the asteroid (and presumably all the other asteroids) looking like stars. Needless to say, there isn’t much you can do with that image.

We were there for 4 nights and semi-fortunately, only got good data on 2.25 of them. The first was clear, the second was cloudy and we packed it in around 1am. The third it was beautiful again and we finished up two of our three clusters, and on the last night, we got just enough to deepen our exposures of one cluster. Of the clusters we wanted to see, the first was up by sunset at ~7pm, while the second only came up at 1am or so and the third a little after that. We just barely got enough for the first two. The first night we got some images of the 3rd cluster (I think) but not in enough different bands to be useful for this project.

Since I mostly work with lots of raw data, as I said, this was my first time literally operating a world-class instrument. If you don’t count outreach activities (I can put together our portable hydrogen-alpha solar telescope blindfolded like the guy in Full Metal Jacket), or TA-ing (using some kind of 1843 brick observatory full of leaves and ghosts to show undergrads that Saturn actually has rings), my experience with these kinds of telescopes was on the back-end, dealing with the imagery itself. It was somehow surprising that to take images you simply type a command into a little unix terminal window and hit “enter.” The first few times you do it you think “that’s it?” It doesn’t seem possible that this is all it takes to photograph these enormous galaxies billions of lightyears away from Earth (and simply odd to have a typed command control a physical object), but of course, years of work went into designing everything and writing the software and wiring the control system… it’s just a thing that is easy to forget in the moment.

Although the operator controls the telescope pointing, this paddle could just move the telescope over a few inches at any moment if I hit it. Good to know it was powered up and plugged in. Other than the possibility of ruining an exposure by accident, the work area was actually pretty nice. Non-astronomers who I mentioned the trip to all assumed that we were exposed to the elements somehow—that we were actually inside the open dome structure. I’m glad that we weren’t, since other than stray light getting near the mirror, it was probably about 30 degrees up there, with extremely forceful wind. Astronomy would be brutal if we had to work that way. Instead, we get to stay in a nice, warm, protected control room. Bonus points for the controls on the operator’s side of the room looking like they could have come out of a 1950’s sci-fi movie with gauges and backlit buttons.

After getting enough rest following the first night I set out to wander and look at all the other instruments. There was some slight awkwardness based on the fact that Kitt Peak is enough of a tourist attraction to draw people during the daytime, and though I was there as an astronomer, it isn’t like I’m carrying some “I’m supposed to be here” card that would make me feel less odd about crossing “Staff Only” signs. I mean, I did do that, but I felt weird about it.

We have a strong involvement in the One Degree Imager (ODI) at the WIYN telescope so I headed over to give it a look. Gravitational lensing studies are all about trying to get accurate, resolved, shapes of galaxies, and seeing is often the most important part of doing that. Seeing is defined as the size of stars in the image—stars are so far away that they should be points of light, but the atmosphere spreads them out, and the larger they are, the worse the image is. The adaptive optics in WIYN’s mirror, and the further corrections made by ODI, (which was only installed in the past year), takes those star images in real time and predicatively corrects the atmospheric distortions. So even though it’s not located on top of a giant windy pedestal, it’s a pretty good tool for resolving galaxy shapes. Even though it isn’t a tourist area, I figured that since I’ve been hearing about it for years and helping to write proposals involving it, I was within my rights to barge in and look around. The operator was kind enough to show me around, and despite the fact that it is a comparable instrument to the behemoth I was working in, the entire building was about the size of a large house. Here’s a picture of the adaptive optics behind the mirror that shape it to improve the focus.

And ODI is this stuff:

Apparently the instrument is so quiet when it’s moving that they actually put a cowbell on it so the operators could be sure that it was actually changing position. It was over where the ODI equipment is located, but so essential that they evidentally built the new instruments around it, so you can’t actually see it anymore. (I assume that it just happens that they were able to put all the new stuff around it, but I prefer the idea that they considered it so indispensable that some planning went into keeping it there). Oddly, almost all the pictures I tried taking of it came out blurry.

As you can see WIYN clearly creates some special zone of optical clarity for itself that transfers blurriness to all nearby imaging devices.

One of the most visually striking things up on Kitt Peak is the unusually shaped McMath-Pierce Solar Observatory. In terms of astronomical instruments it’s certainly unique. Like an iceberg or the hostility among players on a high school lacrosse team most of it is hidden below the surface. It sends light from the sun far down a long tunnel burrowed into the mountain, then reflects it back and finally sideways into an underground control room. At the time it was built this was the best way of getting magnified images and detailed spectra from the sun. Here’s a schematic:

Not knowing anyone in this field, I resisted the urge to actually go in and see the science area, but I did check out the little tourist booth where you can see the shaft and mirrors and took a ton of artsy photos of this futuristic dystopian setpiece.

Having done my share of aimless wandering and rested up, the rest of the trip was pretty fun. Unfortunately, the nights when we were able to get data alternated, so the second night we got nothing, the third we got a lot and on the last a storm was approaching so we only got a few hours done. There were points where we just watched the monitor as clouds were passing in front of the guide star, pausing it when it was covered, opening the shutter again when it cleared up, like the world’s most mundane video game.

One of the nights it was cloudy I went wandering around the building to check out the old equipment and aging rooms. It seems as though when it was designed, they envisioned using the building for everything, and since it’s so enormous they have all kinds of spaces there. People kept mentioning that there are creepy dorm rooms somewhere, in a way that implied that although they worked in this building, they weren’t clear on exactly where they were. Since there aren’t any windows, these terror rooms were basically small prison cells—unfortunately I didn’t find them. This is the exercise room, you can see the angular outside wall behind the state of the art exercise stuff:

They also had a bit of stuff left over from the era of photographic plates. Fortunately for us, we now use CCD cameras, but when this telescope was inaugurated it still had the old-fashioned cameras. So there were instruments for measuring sizes of objects and the astrometry between them on the photographic plates, as well as cabinets full of data, back when our data was a physical thing.

Going outside you really do feel like you’re isolated on top of some kind of desert ocean up there. You can’t see any signs of civilization until the nighttime, when little towns in the desert light up. I haven’t spent much time in deserts and the beauty is so striking.

In conclusion: telescopes.