Caltech

That's all, folks!

And I made it! Things are basically wrapped up for the summer. Sure, it's been *really* busy at times, but that's better than being bored, right?

I finished making the signal to noise peak counting statistic work out, and I completed the chi square minimization on both the peak counts and correlation functions. I ended up implementing a different method of estimating the covariance matrices for the correlation functions, which I was able to use in the chi square minimization. Turns out this showed that the previous method of estimating errors was a bit small.

But, some of the code I had to run at the very end was very computationally intensive, but had a parallel component, so I ran it on the supercomputer my group has a project on. It definitely ran a lot faster than it would have on my work computer!  There was some paperwork to get my account under the group project, which I'm not particularly enthusiastic about, but c'est la vie.

So to finish up your SURF, you have to write a nice final paper summarizing all the work you did and your conclusions. I recently finished writing that, and it's nice to have that out of the way. But after that, I'm giving a final-final presentation at SURF seminar day on campus in October.  For this one, unlike the August presentation, I should actually have some complete results to include, which I am quite pleased about.

After that? Well, I'm happy to say that this is not the end. I'm going to be able to keep working with my group during the academic year. The first goal is to finish up my SURF project analysis, and hopefully publish the results. After that, there is another project where I could do some similar analysis on a different data set from the Subaru telescope.

But this summer's SURF adventure is finished. It was a lot of hard work, and certainly was frustrating at times, but it was incredibly rewarding. I had the opportunity to meet new people, learn a lot of new science, and try out a different aspect of what goes into work in astrophysics. I very much enjoyed it, and because of this summer's experiences, I am thinking that this (astrophysics, cosmology) is something I might want to pursue further in graduate school. And I'm glad I was able to learn that now.

It's been tons of fun, but c'est tout!

Seminars, talks and posters

Sometimes working on my SURF isn't all about work.  It's about learning! Specifically, I have had opportunities to sit in on talks, journal clubs, and presentations. This has included the SURF presentation session on 21 August (where I gave a presentation on my work over the first 8 weeks), and will include the Post-Doc Poster Day (if it is rescheduled before I have to leave for the summer...).

But just like I highlighted a talk from the KISS Dark Matter workshop in July, I wanted to mention a cool and relevant talk I attended a few weeks ago. Ali Vanderveld, one of the postdocs in the group, gave a talk titled "Testing gravity on cosmological scales with weak lensing" as part of the JPL Postdoc Seminar Series.

Ali started out by briefly explaining what (gravitational) lensing is (as this series of talks is aimed at a general audience), and how lensing is useful in dark matter mapping and constraining cosmological parameters. Of course, now we know about 'dark energy' too. Since it is very likely that measurement errors explain the observed acceleration of the expansion of the universe (the evidence for the existence of 'dark energy'), that leaves the culprits that cause this acceleration down to a couple of main ideas. These are a 'dark energy' that is a novel type of matter that has negative pressure, or *modified gravity*. This topic -- modified gravity theories -- was why I was looking forward to this talk. Just the week before I had realized I didn't know much about them, and if I had a bit of free time I was going to ask someone to explain some of it. But Ali was going to give a talk about it, so I was in luck!

Modified gravity theories seek to address the possibility that gravity has different behavior on very large scales (closer to the size of the universe) than it does on small, local scales (like on Earth).  Essentially, these theories explore the possibility that General Relativity is correct for the small scales, but that it isn't correct for the large scales.

Ali went on to discuss how weak lensing measurements could be used to discern if modified gravity really does explain the observed behavior of the universe.  The way to do this would be to look at parameters that vary from theory to theory (in relation to each other and to GR). Then, it would be possible to use observations to place limits on these main parameters, and thus narrowing down the field of possible modified gravity theories.  This has been done in the solar system, and these results show that GR describes the local scale gravity very well.  But it is possible to compute observable quantities from weak lensing observations that can be can then be linked to the theory parameters.

So, with better data, people can start to investigate which (if any) modified gravity theories match observations. But they have to get the observations first. So Ali finished up by talking about the scope (and promise) of future dedicated weak lensing surveys.

It was an excellent talk, and I found it quite useful. I walked away with a better understanding of the scope of the problem: what modified gravity theories are, and how (especially using weak lensing) they might be accepted or ruled out.

Now I get to look forward to seeing posters from quite a few of the post docs in our group!

Station Fire

Just a quick note about very relevant current events. Last week on Wednesday, a wildfire started a few miles away from JPL, up the Angeles Crest Highway.  This 'Station Fire' has since grown to be *huge*. The main JPL campus (on Oak Grove) was closed over the weekend, and the fire line got quite close to here over the weekend -- within a mile or so.  But that part of the fire boundary was contained, and the active burning is not so close to the main JPL campus anymore, so it's out of direct danger.  Still, the Oak Grove part of the lab was closed for most people (like me) on Monday because of air quality concerns.

Things went back to normal here on Tuesday, but the air quality is still pretty bad.  I'd estimate less than a mile of visibility this morning, looking to the East.

But that doesn't mean that things are over. The current estimated boundary size is 140,150 acres (over 200 square miles!), and while JPL is safe, other areas are being threatened.  This includes residential communities, as well as the Mt. Wilson communications center and the observatory. There has been a lot of work that has gone into protecting Mt. Wilson, from laying down fire retardant to back burning to get rid of fuel that the wildfire could use to jump to the facilities on the mountain top. Apparently things are looking somewhat up as far as keeping the facilities safe, but it's not over yet.  The towercam on the 150 ft solar telescope went offline yesterday afternoon, so we can't directly see what's going on up there.

Still, if you're inclined to follow what's going on, the incident website is here.

Photo of the smoke plume, as seen from Caltech on Saturday morning (29 Aug 2009).

Photo of the fire as seen from Caltech Saturday evening (29 Aug 2009).

Tour of JPL

On Wednesday, 12 August, during week 9 of the normal SURF period, Varoujan led the Astronomy summer student coffee group on a tour of JPL.  Why so late? Well, at least one of the viewing galleries we were going to visit was closed prior to that, so late is better than never! And at least for me, my time out at JPL only led me to visit a couple of buildings, and none of the ones with the cool viewing galleries. So I was excited to see new things.

We started out by walking down to the Mars Yard viewing gallery, where work with the Mars rovers is going on.  In one half of the area, parts of the Mars Science Laboratory (MSL) engineering model (especially its legs!) are staying while they are worked on.  The other half is where there's a setup rigged up to try to 'Free Spirit', one of the Mars Exploration Rovers. It got stuck in some soft soil about three and a half months ago. Because they can't exactly send someone to dig it out or fix the rover if they break it while telling it to get itself out, it is worthwhile to spend time studying how Spirit is stuck, and test out methods of getting it out using a rover model and a mock up of the soil situation.

MSL engineering model rover in the Mars Yard.

The 'Free Spirit' setup.

Then we headed over to the high bay in the spacecraft assembly facility, where the spacecraft that the MSL rover will go into is located. In the viewing gallery area there was information about the MSL spacecraft and rover. The plan for landing the rover on Mars is interesting, if not a bit convoluted. Take a look here (the JPL MSL timeline page). Also, the rover has a laser to *vaporize rocks* on its head. I know, there are very good scientific reasons for doing this, but that's just too cool.

First picture of the mission patches on the wall in the high bay.

Second picture of the mission patches on the wall in the high bay.

Most of the pieces going into the MSL spacecraft were covered up...

It was really cool to see all the panels on the walls that show what missions have been assembled in this building. Apparently it was quite a trick to get Cassini out of the door on the left side of the building. It's a really big door, but Varoujan was telling us how it was a very tight fit nonetheless.

Then we wandered into the lower floors of the building I work in. We briefly stopped outside a clean room that had this sign on the outside.

Apparently any lab that has this on the outside contains something that will fly on a spacecraft sometime in the future. I then realized that a different lab that I walked by daily was working on flight hardware... goes to show that there's a lot about JPL that I don't know.

Then we headed up to the Multimission Image Processing Laboratory.  It's funny how I (and probably a lot of people my age) take for granted the nuances and difficulties of data transfer and image processing... and we don't stop to think how people had to work out all the minute details sometime in the past!

Of course, our main stop in the Image Processing Laboratory was the media viewing room.  We saw animations of a number of missions, and videos compiled of images taken on real missions. We even got to wear cool 3D glasses for some of them! It was really impressive. We got to see what the sun looks like over time, with sun spot variation and solar flares. We also saw animations as well as actual footage of Mars and the Martian terrain (martain? I know it doesn't work, but the word 'terrain' seems... out of place).

We finished up by visiting the displays by the Von Karman lecture hall.

A very scaled down model of one of the Deep Space Network satellite dishes.  The building next to it is at JPL, and is an eight story building (iirc).  Sure, that's not the bigest building ever, but the dish *moves*.

A model of the first US satellite. To get an idea of how small that is, those are fairly regular fluorescent light fixtures.

A model of Galileo. The mast extends overhead and to my right... but it was too long to fit into the picture.

Granted the tape recorder is space hardened (for fidelity), but still.  It's interesting how so much information can still be stored on and passed through something that seems so .... small .... compared to today's standards!

The Voyager model that became a movie star! (V'ger in the first Star Trek movie.)

Close up view of the plate placed on Voyager.  It explains how to play the accompanying data disk, has Earth's location relative to the nearest stars (the starburst figure), as well as a model of molecular hydrogen.

A very scaled down version of Cassini.

C'est tout!

Whirlwind tour of life

Just a short note for now!  We're coming down to deadlines.  For SURF, the JPL student seminar days are this week Wednesday and Thursday.  I'm giving a presentation on the eight weeks of work I'll have done so far. (It'll be really good practice for presenting again in October at the Caltech SURF seminar day!)  I'm keeping plenty busy on my work, but I want to talk a bit more about that in another post.  I have three posts rattling around in my brain, but I have to put them on hold for a little bit.  See, a lot of my time away from JPL and my SURF is spent working on the little t, the Caltech undergraduate student guidebook.  I'm the business manager, as well as an editor (okay, the graphics editor. Not much besides that).  And we have our publishing deadline this Friday.  Our whole team (Andrew Price, Amy Lam, myself, and some of our friends who have graciously offered to help) is working hard to get everything finished on time.

So when I've survived this week and these deadlines, I have more tales to share. I promise they're worth the wait!

Second Half of SURF

Well, I'm halfway done. At least with the time for the summer! I think my work is progressing nicely, too. I know, by saying this I am tempting Murphy's law to take effect. And then I'd have setbacks and problems galore, but I'm risking it. Hopefully everything will keep going as planned. Yes, I'm doing work and that involves challenges, but my "extra" problems have been minimal so far.

That's not to say they don't happen. Early last week Joel (my co-mentor) told me there was an error in the original galaxy catalog I'd been given. He had a new copy to give me by Tuesday morning. Of course, this meant I had to rerun everything for my correlation functions. But it wasn't as bad as it sounds. I didn't have to change any of the code I wrote or modified to do this the first time, so all I had to do was rerun the code. Sure, that takes time, but I could do other things while that was running. So I think that qualifies as a minor setback. And now I have shiny new data!

As far as work goes, I'm beginning to look at signal to noise peak counting on the mass map, as determined from the shear data from the galaxy catalog. I'm also looking at the chi square minimization to find the best fit of cosmological parameters based on the observed two point correlation function.  There are some problems with the chi square minimization right now, but hopefully that has a simple solution. Once that's fixed up, I'll get my first results. I'll have to do more things with those values later on, but I'm looking forward to having "real numbers" come out of my work.

So things are progressing nicely, we have our abstracts due today, and that's where things are in my SURF.

Trip to Mt. Wilson

Monday evening was time for a field trip! Varoujan Gorjian and Tim Thompson, who hang out at the summer astronomy student coffee breaks (and do their parts to make them happen at all!), led an excursion up to Mt. Wilson. A few cars of people drove up, and then Tim was kind enough to lead us around. He's trained as a tour guide up there, and he has keys to a bunch of the telescopes, so we got to see lots of stuff.

We started out by looking at the 150 ft and the 60 ft solar telescopes. The 150 ft telescope operator was off at dinner when we first got there, so Tim went over some of the historical importance of these telescopes. Initially, Mt. Wilson was a solar observatory, and between the two telescopes there are the discovery of the solar magnetic field, and the oscillations of the sun, among other things. But they're still in use, as we got to see when the operator got off dinner. He took us inside and showed how a scan was in progress, and explained that while there is no real sunspot activity right now, they still make observations (i.e., drawings) of sunspots by hand when they are seen.

Looking up at the 150 ft solar telescope.

A scan of the solar image in progress.

Then we went over to look at the 60 inch mirror telescope. We got to wander around the telescope itself, look at (but don't touch!) the controls, and finally go down to the base, where the gears to turn the telescope are. They put everything together with rivets back in the day... and the wrench to turn the ones down there is huge, but apparently tiny next to the 100 inch telescope base rivets. (We couldn't see the base of the 100 inch telescope since that's buried underground.)

Base of the 60 in telescope.

Then we briefly wandered off to the south. We got to see some views down into the LA basin, but it was a bit smoggy, so we couldn't see to far. Apparently on good days you can see Catalina Island...  But we did get to see this plaque on a concrete platform. Not some random one, but one with historical importance: this was the platform off of which Michelson performed his speed of light measurements between Mt. Wilson and Lookout Mountain.

Michelson's concrete slab and its more recent commemorative plaque.

On our way back through the main area, we stopped and looked at one of the Chara interferometry telescopes.

Part of the Chara Interferometry Array.

Then we went on to the 100 inch telescope, which actually has a name: Hooker Telescope. Part of the bottom of the telescope was open enough that you can shine a light in and see the base of the piece of glass that was turned into the mirror. It's a green that's reminiscent of wine bottles, and for good reason: French wine bottle makers were best equipped to pour large amounts of glass, so that's who made the piece of glass for the project.

Bottom of the 100 in Hooker Telescope mirror.

We went up another level to look at the controls for the telescope, and Tim gave us a demonstration of how the dome surrounding the telescope moves equatorially, and it was really impressive. It was really hard to tell which part was moving from the inside. The mechanism to move the dome was so smooth that there was no "bumpiness" in the turning. He opened up the door to the catwalk around the outside of the dome, and finally we had an outside frame of reference to appreciate that yes, the dome is moving, no matter what your eyes try to trick you into thinking inside. He was going to just move it a bit to demonstrate, but when he went to move it back the other way, the controls weren't working in that direction. So he had to turn the dome all 360 degrees in the one direction -- and we got to go on a merry go round ride! I walked around the catwalk once, went inside to peer at the telescope versus the moving floor, and back out to just watch some of the sunset while we were turning it back to its starting position. I have to say, that was really fun.

Sunset as seen from the catwalk of the 100 in Hooker Telescope dome.

The 60 ft solar telescope is on the left, and the 150 ft solar telescope is on the right.

Then it was time for us to be heading back. So we piled back into the cars. Let me just say this of the drive to and from Mt. Wilson: it's windy. Very windy, with narrow shoulders and very low speed limits. I thought it was cool, and a very scenic drive, but I can appreciate how some people can get car sick from that trip.

So Tim dropped a car full of us back at Caltech, and the day was done! My first trip up Mt. Wilson was a success.

Complications are a part of life

Sometimes things go right, and sometimes there are problems. This week is one of the weeks with problems, and thankfully they aren't too bad. 

So I was working along on Wednesday, having run the programs that uses chi square minimization to compare the theoretical and measured correlation functions for the first time. I had to tweak a few things in the constraining programs, so I was expecting to have made at least a few mistakes that I'd have to fix. But I still got data out, so I could work with that. So I begin trying to figure out what I could be doing better, because the first run wasn't quite right, when my IDL license server can no longer be reached. So my IDL sessions quit, meaning I couldn't really work on much of anything. Sure, I could read through things, but I couldn't test anything.

It wasn't my fault at all, and it was a fairly easily fixed problem, but it was still frustrating. I'd been lucky so far and little things didn't go wrong, and thankfully this was truly easy to fix. The problem was cleared up (as much as possible) by about 9am on Thursday.

So for the rest of this week I've been working with the constraining code. It's progressing, but I'm definitely making little mistakes all over the place. And the symptoms of the errors are sometimes really bizarre, but thankfully it only takes about an hour or so to rerun these routines, not more. So yeah, some frustration, but I know I should be a bit more optimistic, because I am still moving forward. But you have to do what you have to do, and sometimes that's debugging.

Summer research = reports

It's the end of the third week of SURF... which means it's time to turn in a progress report. As part of the 10 week program, you have to write two progress reports, an abstract, and a final paper. You also give a final presentation (or poster) at the end of the summer (there's the JPL SURF seminar day on 19 and 20 August this year) or on campus in October. It's been my experience that the progress reports aren't too much trouble. You go through what you've worked on, what you will work on, and what challenges you've face, and that's basically it! And after working on a project for a few weeks, you get to know it fairly well, so it isn't so hard to talk about those details. So for this progress report I'm fairly relaxed, since my project has been going smoothly! Just a few more edits, and getting it signed, and I'll turn it in.

The highlights of my progress so far are modifying the correlation function measurement code to include redshift data (ie information about distance), running this code on my data set, and beginning to compare the newly measured correlation functions with theoretical ones. (I'm using previously written code to calculate the theoretical correlation functions). I need to play around with varying parameters on the theoretical calculations, because right now the theoretical curves don't quite match the measured data points. This might be that I haven't varied the parameters in the right way yet, or I made some silly mistake, but I do hope I can get it figured out quickly. Then I can move on to the next part of my project.

It's a brief entry, but that's what's happening right now. There will be more later!

Dark Matter

Third week already! Where does the time go? (Okay, I know where it goes, but I'm having lots of fun so the time is passing quickly.)

So what's new up here? Well, there won't be any group meetings for the next two weeks. That's because there is a dark matter workshop happening at Caltech for the next week, organized in part by Lexi (Leonidas Moustakas). The workshop, 'Shedding Light on the Nature of Dark Matter,' is a mini study program being run under the Keck Institute for Space Studies (KISS), which is an institute under the guidance of Caltech and JPL. Quite a few of the group members are active participants in the proceedings, and even more people will sit in for some talks and discussions. So it isn't like there won't be tons of interaction happening when there are no group meetings.

Andrew, Faith and I went to the opening session and talk on Monday morning. We're all working on different topics and sub areas in our projects, but a) dark matter is relevant for lost of aspects of astrophysics, and b) dark matter is really cool. There is a workshop happening in our backyard: how could we not sit in for a bit of it?

The introduction talk was given by Joel Primack, and aimed to describe the history of dark matter. He covered quite a bit of historical ground, from Zwicky in the 1930s and his observations of galaxy cluster behavior to other observational evidence necessitating dark matter. He also covered some of the work done to try to explain the nature of dark matter, mentioning the hot dark matter, warm dark matter and cold dark matter models. (The simplest difference between the three is the velocity dark matter would travel at, hot being ultrarelativistic, cold being nonrelativistic, and warm being somewhere in between the other two models. Based on what is now known, a modified cold dark matter model -- lambda cold dark matter -- explains things the best.) But most of is introduction was the history of who did what, and not an intro to dark matter. This makes sense in the context of the workshop, but I'll admit that I was hoping for more basics, since that's the sort of remedial info that I need!

I'm planning on going to another talk in the series that relates to weak gravitational lensing. It's one of the symposium level talks, but that's just fine with me. I am still a newcomer to all this material, so those talks are the right level for me!

But even though Primack's talk didn't cover all the basics of dark matter, I was in luck. I was reading through the astro-ph new postings to arXiv (a really awesome open access collection of e-prints -- check it out!) on Tuesday when I saw a paper titled 'Dark Matter Astrophysics' (by D'Amico, Kamionkowski, and Sigurdson). It was based on lectures some of the authors had given, and were created for the purpose for providing a review for people who don't know much about dark matter. This was exactly the sort of thing I was hoping for, and it popped up at a wonderful time! Andrew and I saved copies, and I starting reading it that morning during breaks from working on my code (as well as over lunch). The final clue that this was just the intro material I was looking for was that Joel emailed me a copy of the paper in the early afternoon, a few hours after I had started reading it.

So what is dark matter? Here are some highlights that the paper covers (but I encourage anyone interested to go to the paper itself -- there are some equations, but for the most part the authors cover the material at a level that is very accessible even without prior knowledge). The initial evidence for dark matter came in the 1930s, from observations like Zwicky's. By the 1970s, a lot of evidence had accrued from observations like looking at galaxy rotation curves (quantifying the rotation of stars and gas inside galaxies) and looking at galaxy clusters, including their dynamics and how light from distant objects is lensed (weak and strong gravitational lensing!). These observations provided measurements of the total matter density in these areas (i.e. gravitationally interacting matter). But when they considered these values along with measurements of baryonic (normal) matter abundances and tried to model how the universe behaved following the big bang, they figured out that there had to be way more matter in the universe than just baryonic matter. So since the 1970s they've known that dark matter is non-baryonic.

Next comes the discussion of what dark matter could be. (This is physics that I'm not too familiar with at all, and I haven't had the chance to really read through this section, so I'll be brief.) So what do people know? Dark matter has to be dark, since it doesn't really emit or absorb radiation. The distribution of dark matter in galaxy clusters -- as a sort of halo surrounding the cluster, in that it is fairly spread out -- means it can't interact too strongly with itself or with baryons. They also know that it basically can't be a Standard Model particle. So what sort of solutions have physicists come up for this mystery? Apparently the best fit so far are WIMPs (weakly-interacting massive particles). Let it not be said that researchers have no sense of humor, especially because the idea of MACHOs (massive astrophysical compact halo objects) was also kicked around as a dark matter candidate (though, to my knowledge, because of the observations that have been made, it's been determined that MACHOs couldn't make up very much of the dark matter in the universe).

Certainly I am omitting some arguments and evidence for the existence of dark matter, and definitely skimping on the discussion of the particle physics, but that's what the paper was written for! I just wanted to share some of the really exciting bits, as I see it. What can I say? Science is cool.