Difference between revisions of "IC 417 Summer visit logistics"

From CoolWiki
Jump to navigationJump to search
Line 36: Line 36:
 
##IPAC. What it is .. and how it is not the Caltech astronomy department. I have no pull at all with the admissions office because I am not a professor!
 
##IPAC. What it is .. and how it is not the Caltech astronomy department. I have no pull at all with the admissions office because I am not a professor!
 
##Basics of star formation
 
##Basics of star formation
#image review for our ~200 special objects - ideally we would get through all of them, but certainly could keep going after the visit (e.g., not on ‘critical path’ to making progress on visit)
+
#image review for our ~200 special objects - ideally we would get through all of them, but certainly could keep going after the visit (e.g., not on ‘critical path’ to making progress on visit). It would be good to at least do some of them in advance if not all of them ...
 
#Basic Excel skills - I added some stuff below -- but this is NOT critical to do prior to your visit - we can and will go through this and learn it as we go. But if you come in knowing this stuff, it will be easier... and you will be in a position to help the others.
 
#Basic Excel skills - I added some stuff below -- but this is NOT critical to do prior to your visit - we can and will go through this and learn it as we go. But if you come in knowing this stuff, it will be easier... and you will be in a position to help the others.
  

Revision as of 17:20, 21 April 2015

High-level schedule

The work days we've agreed upon are June 15-18, traveling June 14, and June 19. There will be minimal help from me on June 19 as sort of a "training run" for when you go home. (It is also a 'safety valve' if we don't get enough done the rest of the time.)

I propose you come in relatively early on June 14, and plan on doing dinner at my house that night.

  • Sunday morning/early afternoon - arrive in LA
  • Sunday night - pizza party at Luisa's ... pizza arrives at 6. come by 5 or 5:30 if you want to play with Andrew!
  • Monday morning 8:30??? - tour of SSC, meet Wannetta
  • Monday morning 9-12? - lectures
  • Monday afternoon 1-5 - get started on data!
  • Tues AM - JPL Tour??
  • Tues PM-Wed AM-Wed PM - work!
  • Thurs morning - Luisa to stay out of your hair most of the day. Work on your own or small groups. Try to do some of the tasks we did as a group. Compare notes. Can you work on your own at home without me?
  • Thurs afternoon - Reconvene for any questions and help, discussion, etc.
  • Friday - return home

Software to install

Make sure you (and all your students who are coming) each have a functional laptop that you know how to use with as much of the relevant software installed as possible well before you get on the plane. Trust me. Makes it much easier if you do all this ahead of time, including starting it up to make sure it works.

  • a web browser (Firefox, Safari, or Google Chrome -- NOT MICROSOFT INTERNET EXPLORER)
  • Excel or other spreadsheet program (Google Docs is ok if your school lets you access it; most of the rest of us will be working with various versions of Excel)
  • ds9 (though maybe you can make do with IRSA tools)

Also make sure you have all the passwords you need for installing new software, getting on a wireless network, or getting back into your machine if it reboots.

Besides the computer, you will need something to write on and something to write with to take notes! Experience has shown that this should be more than post-it notes. I will give you handouts, so the hyperorganized among you may want, e.g., a 3-ring binder.

Pre-visit priorities

If I had to pick priorities for what to have you work on for the next ~6 weeks between now and your visit, I would pick:

  1. math skills (scientific notation, logarithms) - critical to getting through the visit
  2. astronomy skills (magnitudes, HR diagrams) - fairly darn critical, but we will discuss. Just want to make sure you have the basics under your belt.
  3. concepts from the videos that I've dropped into this playlist (still being added to): https://www.youtube.com/playlist?list=PLjCjDYabTFm_5GYwt9A8hRh6OCGPG3csv If you want your kids to watch those too, great; if you want to relay/expand on critical concepts, make sure to hit:
    1. infrared light. There is light beyond the visible, and we will be working a lot with infrared, so you best know the basics.
    2. WISE and Spitzer. Space-based observatories. One is an all-sky survey, one is a pointed mission. Why having both is important. Resolution matters. Jargon for channels/filter names
    3. IPAC. What it is .. and how it is not the Caltech astronomy department. I have no pull at all with the admissions office because I am not a professor!
    4. Basics of star formation
  4. image review for our ~200 special objects - ideally we would get through all of them, but certainly could keep going after the visit (e.g., not on ‘critical path’ to making progress on visit). It would be good to at least do some of them in advance if not all of them ...
  5. Basic Excel skills - I added some stuff below -- but this is NOT critical to do prior to your visit - we can and will go through this and learn it as we go. But if you come in knowing this stuff, it will be easier... and you will be in a position to help the others.

Math and Astronomy to know

Scientific notation.

  • Caroline sent something around in email.

Logarithms.

Logarithms and the magnitude system:

Units (mks, cgs).

HR Diagrams.

Excel to know

Basics of Excel programming. NOTE NOT VISUAL BASIC. just basic Excel. this is what i could find in a brief google search:

Excel skills you will absolutely need:

  1. Reading a text file into Excel. (I do have a screencapture tutorial on that -- again, developed for another team, but fundamentally the same concepts even if the first little bit isn't directly applicable : http://www.youtube.com/watch?v=nCJ3ctOGvNk&list=UUQfN0BTwaSZ6ABsQcieCpdA&index=1&feature=plcp
  2. Convincing Excel how to ignore certain data. In my world, I need a placeholder for 'no data', so in most of my tables, -9.00 means 'no data available'. The most efficient way to convince Excel that that is 'no data' is just a global search and replace. There are some subtle tricks on this, which is why I am listing this as a separate skill. Just changing the range of plotted axes is not enough.
  3. Manipulating columns. For example, programming column E to be column B-D. More handling of "no data" issues here. (For a test set, I could limit this to just be objects where there are data in all columns, making the 'no data' issues non-existent. But for the "real thing", this will be an issue.)
  4. Plotting columns against other columns, and setting limits on what is plotted such that the 'no data' objects don't appear. Personally, I think plotting in Excel is a gigantic pain in the @$$ and I never get it right on the first try. it is far easier for me to write code to make plots, so that is what I do. I will be of limited help on this; I will just end up poking buttons until I stumble on what works for your excel version. You need to pick 'xy plots' or 'scatter plots' or other words like those (exactly what you pick is dependent on your excel version.) There are several you tube tutorials from the general public that I can find that address making plots in Excel. (Again, for a test set, I could limit this to just be objects where there are data in all columns, making the 'no data' issues non-existent, but this will be an issue in our "real" data set.)

Actual files to practice on:

  • Walking (simplest) -- File:Simplest.txt -- set of 20 objects, all known young stars, columns are name, j, h, k, w1, w2, w3, w4, all in magnitudes. File has only actual detections (as opposed to limits or 'no data' flags) at JHK, all four WISE bands. Can you read this in, and say, plot, W1 vs. W1-W4? Remember that brighter is smaller numbers in magnitudes, so for this plot, you will need to reverse the axes so that down is a larger number, and up (brighter) is a smaller number. How about J-H vs. J-K? This involves skills 1, 3, and 4 from the above. Plot whatever else you want; this was just a suggestion.
  • Jogging (harder) -- File:Jogging.txt -- set of 40 objects, all known young stars. Same columns as before. Now has some -9 values, which indicate no data. Same task as 'walking' (make those plots with those issues), but now cope with -9's. Note that for plotting just one band against another, fixing the axes as displayed will remove the -9s. But we essentially never plot just one band against another; it's usually one band against a color, or a color against another color. If you have no data in just one of the two channels, you can fix the axis limits. But if you have no data in BOTH of the two channels, -9 - (-9) is exactly 0, so just setting limits on your plot will not remove those objects. (Note also that there might be some values < 0 that are legitimate magnitudes, but there are no legitimate magnitudes as small as -9.) This highlights skill 2 above. Plot whatever else you want; this was just a suggestion.

If you get this far and think, 'pff, easypeasy', try this:

  • Running (hardest) -- File:Running.txt -- set of 552 objects, with mostly the same columns, but one new one at the end: status. If "status" = 'knownyso', then it is a known YSO; if "status" = 'rejcand', then it is a rejected candidate YSO object, e.g., not likely a YSO. There are all just real detections in here (at least, I think that is the case...). Plot whatever you want, but make the 'knownyso' objects one color or symbol and the 'rejcand' a different color/symbol. Try (just to pick a plot out of the air) H vs. H-K. What is the average H for the known ysos? The average H for the rejected candidates? Can you identify the specific reddest objects in this plot (e.g., largest H-K)? What is the brightest rejected candidate? Draw a line in this plot at H=8 and H-K=1. How many known YSOs are in each quadrant of that plot (eg., above and below H=8 and left and right of H-K=1? (This is pretty advanced skills.) The Excel functions 'average' and 'countif' may be of some help in doing this.

That oughta be enough to keep you busy... :)