First light for BICEP2

Hello y'all. My apologies for not writing in a week. Let me catch you up. This past week Ross the winter-over from last year came in for a two week stint to catch up this year's winter-overs Dan and Dana on all of the things that they did last winter. Ross is british and very good with computers. Last winter he wrote a very good collection of programs that deal with automatically processing the data from SPT. For those of you who are interested in that kind of thing, I'll describe it briefly. The crontab job calls several bash scripts every couple of hours. The first bash script updates a Mysql database which keeps track of numerous meta-data things including which fields and sources in the sky have been observed at what times, and what of that data has been processed. The other bash scripts collect the appropriate arguments then call python scripts to process the data. These python scripts are really just wrappers over underlying IDL procedures, and they interface with the Mysql database to figure out what needs to be processed and collect arguments. The IDL jobs are then collected and sumbitted to a Sun Grid Engine, which handles queueing jobs, making sure the processors do not get overloaded, and collecting the output. Once data is processed, the output is written and the Mysql database is updated. It is pretty cool and works well, and it all makes sense once you wrap your head around it all. Ross is coming back to Chicago after being at pole to work as a post-doc for Carlstrom and the SPT group.



In terms of work for myself this past week, I have spent a fair amount of time looking at Ross' autoprocessing procedure to understand how it all fits together. One hinderance is that I have not worked with Mysql at all before, and have only written one program in Python 5 years ago. So I am learning those languages as I go. But as my uncle says, once you know a programming language well - which for me is C++ - you can pick up enough to understand programs in other languages pretty quickly.

Besides that, most of my time has been spent writing the new scan and schedule files for the next year's CMB observations. These are the files that tell the telescope where to point and how to move. The files are pretty high-level, meaning that there are already a bunch of functions that take care of the details and I just invoke commands like "acquire($source)" which tells the telescope to track a given source that is defined in a configuration file elsewhere. We will test out my programs at the beginning of this week.

Beyond my immediate work, the most interesting thing that happened this week was the BICEP science lecture last monday. As most of you know, I worked on BICEP2 last year, and they have been deploying the new instrument this season. Things are going well for them, and at the lecture on Monday they presented their first CMB field maps. It is pretty incredible to see how everything comes together to produce maps in the end. I think back to one year ago when I was still struggling to understand the gcp (Generic Control Program) so I could get it to talk to the detector-readout electronics. I remember how incredulous and excited I was the first time I saw data from the detectors actually propagate all the way up through the gcp and show up in plots in the Viewer in March of last year. Then I remember fighting data-frame packing last summer, which is required to match weather and position information from the Antenna side with data from the detector side. To see it all come together is pretty fantastic, even though I haven't worked on the project since September. It was especially cool to see the first BICEP2 CMB field showing temperature anisotropies that had been measured in 20 hours of observation side-by-side with the same patch of sky as measured by WMAP over 5 years - they looked virtually identical. Now this is not a fair comparison because WMAP had been measuring the entire sky in those 5 years where BICEP had spent 20 hours on that small patch, but it was still cool. The talk was dedicated to Andrew Lange (see last post).



Finally, the week in recreation! Did I tell you all that there is a climbing gym here? Well, there is. It is a small bouldering cave, but it is enough to have fun and stay in shape. And there one guy, Orin, who is climbing stronger than I am and puts up problems that I can make progess on but have to work at a lot. That is totally awesome, having people around climbing harder than you, which makes it easier to push yourself. I put up an awesome problem last weekend, and was finally able to send it this Saturday. The crux went from this small crimp to a spok-grip on a sloper to a thin pocket... Hmm, I need to include a beta video on this blog! Also awesome, Orin has time off in NZ at the same time that I do, so we are planning on climbing together for two weeks there. Hopefully some long rock routes and maybe some mountaineering stuff!

I've also been playing a bunch of volleyball and some soccer. We played volley ball for almost 4 hours on Friday evening! Soccer is a killer on the lungs, so less people play. Last Wednesday there were only 4 of us, so we played 2 on 2. Brutal! At this altutude, we played for 10 "quarters" followed by breaks - that was all that we could handle.

Every Wednesday is Pub Trivia. There are usually 4 rounds, and the winners of each round get a 6-pack of beer. It works out well because I don't know hardly any of the questions, but I don't drink our team's beer when we win either! Lots of pop-culture questions about movies and popular music, with some random questions like the round on robots last time. This coming week the SPT team is hosting, and I am putting together a round on mountains. I'll post my questions here, and see how many you guys can guess!

Back to work on this cloudy Monday morning. I hope that all of you have a brilliant day!