Monday, June 28, 2010

Unsuccessful and Unsatisfying

I usually post good, happy bits, but I'm not feeling too hot about today.

Goal: write some code to take a file with info on 4 million+ stars and create a function that relates apparent magnitude to magnitude and color errors. This will then be useful when generating more accurate CMDs.

I did start this, and eventually created another file that only used data of 500,000 stars so IDL would run faster. However, I spent quite a few hours working with the 4 million star fits binary table, so a lot of my time was spent waiting (and waiting, and waiting) for IDL to give me a different error message. Not a great feeling. Also, my for loop is irky, and I don't feel really experienced enough to debug it properly.

Tomorrow morning hopefully I'll figure this out. Especially because our group meeting tomorrow marks the halfway point for research and we will share our progress from the last 5 weeks, and goals for the next 5 weeks. I've been really good about bringing things (figures) to group meetings, but I'm afraid my rut will prevent me to bring my usual good news and progress...

Friday, June 25, 2010

Quick Friday Update

Today officially marks the halfway point of this summer research extravaganza!

I was not nearly as productive as I would have liked to be today, but I did get some things done. Mostly, I worked to write functions that create a fake CMD given a certain isochrone. I eventually got a product, but it took far too long, in my personal opinion. Gail and Mimi both headed out early, and so instead of persistently asking questions, as I typically do, I pushed through and figured it out more or less independently. That definitely got frustrating at times, but I feel like I have a deeper understanding now of how to write functions in general.

So, hooray for the weekend! In between lines of code for the past few weeks, I have been creating the greatest Summer Music Mix that has ever been put together. Today I finished it, so if any of my many millions of readers out there want a copy, just give a holler.

Thursday, June 24, 2010

Fake Progress

I mean, it's real progress, but my fake dwarfs are progressing quite nicely.

Yesterday, I did some more IDL learning, and wrote functions generating star positions in fake galaxies that fit an exponential, circularly symmetric model. Today, I wrote a function converting these positions into x and y coordinates in terms of exponential scale lengths. Beth then gave me a little bit of contour-plot bootcamp, and she helped me produce this little number:

... nice, huh?

This afternoon, I worked on creating a function to convert these x and y coordinates in units of exponential scale lengths to angular size in units of arcmin, as well as plotting an isochrone of an old, metal poor galaxy, as provided to me by Mimi. As I write these functions, I'm also trying to keep track of my work through headers that explain what the function does and how to use it.

I've really been enjoying these simulations. A week ago, this blog post would have seemed like a foreign language. But now, I'm excited to work more and more on this project of creating these fake galaxies, mostly because I know where this work is heading. Soon enough, these fakes will be put in simulated fake backgrounds, so we can determine their detectability. And, ultimately, we'll determine how detectable dwarfs like these fakes would be in our own sky...

As a side note-- a special blog Thank You goes to Beth, who has been putting up with me in her office for long periods of time these past few days. I know she has a ton to work on, so I'm feeling pretty grateful that she takes so much time out of her day to help me. THANKS BETH!

Tuesday, June 22, 2010

Making Fakes

It's happening! After closing loops on old data analysis projects, and writing about it LaTeX, I'm beginning the process of making fake dwarfs!

What this means, is that I take random star samples from the Ursa Major HST stellar data set, and make them into other "fake" dwarf galaxies. I did this many times, and had not 1, not 2, but 25
figures to bring to group meeting today. They were all of CMD's of these fakes, some of several fakes of the same sample size overplotted on one another. After group meeting, I calculated total magnitudes of these simulated dwarfs.

Next step: randomly distribute said stars spherically with an exponential fit. Beth will show me how to do this tomorrow.

Big goal: put these fakes at different distances and against different backgrounds, and LOOK FOR THEM!

We're approaching the halfway mark soon enough, so it would be great to get hopping on this!

Science is so cool.

Friday, June 18, 2010

Something to show for the past four weeks

I have the most beautiful LaTeX document that has ever been made. Or, at least I think so, (in a maternal way).

Let me tell you a little about it. First and foremost, it documents every piece of valuable work I have done over the past four weeks. It has every derivation, every error calculation, every figure, every investigation... I'm very proud.

Also, all the bits and pieces are in the places I want them to be! I know, I know-- I vowed yesterday not to consume my time with layouts, but I did it anyway. Although, it didn't really take too much of my time. There are apparently plenty of people out there who are frustrated with this flaw of LaTeX, and several of them wrote about how to fix those problems and published their findings on the wonderful World Wide Web.

Also, in addition to the sections I planned to write as I posted on Tuesday, I also wrote a section analyzing the depths of different stellar data sets in hopes of eventually simulating fake dwarf galaxies. By looking at data sets from 5 dwarfs, and a data set from one globular cluster, I was able to determine which would provide the best CMD to model our fake dwarf on. (Turned out to be Ursa Minor, but I'll let you read the document for the reasons why.)

Next week, I'll (hopefully) start writing the code to make these fake dwarfs, either with this data from Ursa Minor, or from simulations from a program called StarFISH.

Thursday, June 17, 2010

Wednesday and Thursday Updates

I've got to keep track of what I'm doing, so I feel like I'm making progress, right?! So, some lists of things I've done, am doing now, and will be doing later. Must be in list form, because at 11:57, my blood sugar is running a little low before the lunch-hour.

What I did yesterday:
- Wrote section (3) of the LaTeX document and inserted all figures. LaTeX sucks at putting figures where I want them. Gail says to just let it be. I've poked and prodded and made some progress, but it still doesn't look great.
- Finished Method 2 for Calculating Avg Surface Brightness at Half-Light Radius-- (That REALLY needs an abbreviation... AvSB@HLR?)-- and then compared it graphically to Method 1.
- Went to a talk by Peter Love about "the foundational questions in quantum mechanics". It was designed to be for anyone doing research this summer, not just a physics audience. It was a great talk. I sat next to my Enviornmental Biologist/Chemist roomate, and she enjoyed it fully, too.

What I've done this morning:
- Put the Methodology comparison figure into LaTeX. Proofread and edited the whole document. Played with page formatting a little more, not a lot of luck having figures go where I want them. (I'm deciding RIGHT NOW to no longer waste time dwelling on making this document look great. I have other stuff I need to be doing.)

What I'll do this afternoon:
-finally get to analyzing different dwarfs (and one Globular Cluster) to see how good they would be as sources to make fake dwarfs for our algorithm.
- put those conclusions into the LaTeX document

In the meanwhile-- Mimi is being AWESOME and decided to take on a calculation Beth requested from France-- something about changing the units of our RA and DEC data. I must admit, it didn't sound too fun, so I'm grateful Mimi volunteered...

As an exciting side note: Beth's using one of my figures in her talk tomorrow in Lyon!!

Tuesday, June 15, 2010

Not only am I an IDL graphing machine...

...I'm also a LaTeX machine!!!!

While Beth is away, I'm working to compile all the work I've done so far into a LaTeX document. I have never used LaTeX before, and I'm kind of enjoying it. I think, like graphing in IDL, I like figuring out the different commands to make my work look good. Here's the outline for my document:

1. Intro
2. Calculations for Average Surface Brightness at Half-Light Radius (both Methods)
3. Figures (Surface Brightness v distance, Size-Luminosity, ellipticity, Surface Brightness Methods comparison)

I was able to get through writing section 2 this afternoon, mostly because this morning, I calculated the Avg Surface Brightness at half-light radius using Method 2-- i.e. using an exponential model and applying it to the known central surface brightness.

Beth wanted me to do this by integrating in IDL, and calling a function I wrote for the exponential relation. However, that didn't work out. Hopefully she'll show me where I went wrong in that procedure when she gets back from France, but I for now I solved for all of the necessary integrals thanks to Now I'm just waiting to input the finished data for central surface brightness care of Mimi-- (who's back from home! hooray!)

Here's what's up for tomorrow:
- use Mimi's data and find Surface Brightness with Method 2
- make figures comparing Methods 1 & 2
- start writing section (3) in LaTeX document

Monday, June 14, 2010

Finished Graphs

... at least I hope they're finished. Still need the A-OK from Beth.

Look at 'em! They look so good! (Click to see better image.) The first graph plots Average Surface Brightness at the Half-Light Radius v. distance from the Milky Way. The second and third graphs are Size-Luminosity graphs. The second graph is more zoomed in, and labels the MW satellites. The third graph also plots Globular Clusters.

Thursday, June 10, 2010

Moving on to Phase Two

After almost three full weeks of working in IDL to organize and analyze the known data about previously discovered dwarfs around the MW and M31, I'm getting ready to move towards creating an algorithm to find undiscovered dwarf galaxies.

To kick-start this process, Beth had me read bits of a paper by Walsh, Willman, and Jerjen entitled "The Invisibles: A Detection Algorithm to Trace the Faintest Milky Way Satellites." Basically, the method is to take stars from known dwarfs, and combine them into "fake" dwarfs, that can then be placed behind different backgrounds of stars to create a program to look for similar objects in the real local neighborhood. Who knows! Maybe this could lead to the discovery of the Barlev-Fuchs 1...

In other news, I've more or less tied up all of my loose ends in terms of plots and data analysis on the known dwarfs, and Beth even showed the plot from Tuesday in her brief talk at AstroPhilly yesterday at Swarthmore! It was nice to see a figure of mine projected on a big screen at a talk for the first time! AstroPhilly was good, too. It was interesting to hear about the range of projects going on at neighboring institutions. Also, we got to look at Swat's new 24" telescope. I'm jealous, mostly because currently our dome needs manual pushing in order to rotate.

Looking ahead -- Beth will be in France at at conference next week, and so I'll need to figure out things to keep me productive in her absence. Hopefully that can be panned out tomorrow afternoon.

Tuesday, June 8, 2010

A Graph to be Proud of

Beth wants us to have nice figures to show at AstroPhilly at Swat tomorrow. Lucky for her, I happen to be a beast at graphing.

Here's what I'll be showing (sorry for the poor quality!):

Its the Size-Luminosity plot I've been working on, but with different symbols for MW dwarfs, M31 dwarfs, and Globular Clusters. Also, those dashed lines are lines of constant surface brightness.

Hooray for good graphs!

Monday, June 7, 2010

Loose Ends

Today and tomorrow, my goal is to tie up all of the things I've been working on for the past weeks.

After meeting with Beth this afternoon, we made a giant to-do list of things that need to happen before I can close up these projects. Here are the projects I have, and the things that need to be done to them before they're finito:

1. Calculating the average Surface Brightness within the half-light radius. Beth and I spent a long time discussing the effect of ellipticity on the half-light area, and thus the Surface Brightness within this area. I've since made a few pretty figures that shows this relationship, both in general, and with respect to our specific data. I'm still trying to figure out if I can make the assumption that the Elliptical half-light radii calculated in the Martin et al 2008 paper can be used for the major-axis radii of the half-light ellipses. (Gosh, that was a figurative "mouthful" that I hope I still understand tomorrow...)

2. Graphing distance from home galaxy vs. said Surface Brightness above. So far, I have this for Milky Way dwarfs, and the graph is labeled and pretty. However, some error bars are missing, so I need to account for those tomorrow.

3. Graphing size vs. Mv, with lines of constant Surface Brightness overplotted on top. Still need to do that last step. I think I have all the pieces to accomplish that tomorrow.

Great. So, hopefully, all of this can be finished tomorrow-- (unrealistic?)-- and then move on to bigger and better things. That way, I can struggle with those bigger and better things while Beth is away in France next week...

Friday, June 4, 2010

nothing FITS better!

An update from yesterday morning that I am only now writing about:

The FITS Binary table is complete! Or, if not "complete," at least up-to-date with all the data we've been working with. I successfully structured and stuffed, so now our data will be easier to manipulate. I think Mimi's already working with it to compute each satellite's distance to it's "home" galaxy.

So, Hooray! Successfully completed. And it was fun, too, surprisingly. I suppose part of this learning process is also learning how to enjoy the day-to-day tasks that I know will ultimately help the big-picture project.

Wednesday, June 2, 2010

A Quick One

OK, just a quick post for mental organization.

Beth showed Mimi and me the wonderful ways of FITS binary tables today, and I'll be structuring the one we'll be working with. I left Beth's office thinking-- great. I completely understand what I need to do, and I'll do it ALL before I rush home to play Ultimate in Philly.

Of course, that never works, does it? There were a few alignment issues in the ascii data table I needed to read in, and so I spent the last hour editing and lining up columns of data. So, at least that nitty-gritty step is done.

Up next: structure the FITS binary table. I'm ready to come in tomorrow morning, pump some tunes, and knock it out. Hopefully, once we have that set up, it will be easier and neater manipulating our data.

Tuesday, June 1, 2010

I'm an IDL Graphing MACHINE!!!

...or something like that.

Over the past few days, I've been learning how to make detailed, attractive graphs of the huge dwarf data table we're working with. I now know how to make graphs in color, have error bars on data points, and even label the data points accordingly. I made a really nice graph today of Distance versus Surface Brightness for all of the known Milky Way dwarf companions. (I wish I could post it here! I'm extremely proud of it.) I spent a lot of time labeling each dwarf, so the graph was readable (dare I say, beautiful?!)

Also, graphing allowed me to catch a pretty important mistake, today. While plotting Size versus Mv, I noticed some of the Magnitudes were positive, when they were all supposed to be negative. After close inspection, I realized that the way I read the data table into IDL was incorrect. Every once in a while, it would cut off a character from the beginning of the string in a column. I then reformatted, and double checked that every data point was correct. It was necessary, but caused some frustration and eye-crossing. Glad it's fixed now...

Next up: Making that Size vs. Mv graph more sophisticated, with lines of constant surface brightness over-plotted onto the data. Don't know how I'm going to tackle that yet, but I will, possibly (probably) with the help of others in the lab.

Also, as per Beth's wishes, I'm going to post here more. I haven't felt like I've done anything huge or note-worthy lately, so I haven't posted. But I guess that's nonsense. If I'm in lab for eight hours a day, I must be doing enough stuff to write about. I think I've only liked posting in moments of success or excitement so far. So, New Goal: write even when feeling stuck or uninspired. Then, later, when I've worked through it, I'll be able to visibly see my progress.

As an inspirational closing note, I was sent this link last week, and have been looking at it whenever I need a pick-me-up. Enjoy: