Difference between revisions of "C-WAYS Fall work"

From CoolWiki
Jump to navigationJump to search
m
m
 
(15 intermediate revisions by the same user not shown)
Line 1: Line 1:
  
=Sep 6, 2012=
+
=Sep 6, 2012 - big picture and image assessments=
  
 
[[file:email20120906.txt]] - email from luisa with tasks and timeline questions from here on out
 
[[file:email20120906.txt]] - email from luisa with tasks and timeline questions from here on out
Line 12: Line 12:
 
* in any case, we need to get moving on the photometry on the spitzer images in the next 2-3 weeks.
 
* in any case, we need to get moving on the photometry on the spitzer images in the next 2-3 weeks.
  
=SED task - given 9/6/12, due 9/13/12=
+
=Image Assessment task - given July, due Sep 10=
 +
 
 +
There seems to have been a lot of confusion about copying stuff back into the master spreadsheet. [[file:email20120910.txt]] - email from luisa
 +
 
 +
Email from Sep 14 included a 'final' file merging everyone's information and providing a 'final' verdict on every inspected item. Do NOT invest hours in trying to correct or backfill the assessments you're responsible for, because everyone made mistakes, I corrected them where I could, and where I couldn't, well, in any case, we've just moved forward from here.  Outcome of this process was 301 sources, to be passed to SED assessment task, below.
 +
 
 +
 
 +
=SED training task - given 9/6/12, due <strike>9/13/12</strike> 9/20/12=
  
 
here is the list of SEDs to make more or less by hand.
 
here is the list of SEDs to make more or less by hand.
Line 19: Line 26:
  
 
guidelines:
 
guidelines:
*start with your notes from july and the full catalog i gave you (at least i think i gave you) that has ALL the bands currently in my master DB, optical through 70 um. if you can't find it, it's still here:
+
*start with your notes from july and the full catalog i gave you (at least i think i gave you) that has ALL the bands currently in my master DB, optical through 70 um. if you can't find it, it's still here: http://web.ipac.caltech.edu/staff/rebull/working/tuesAMsourcelists/
http://web.ipac.caltech.edu/staff/rebull/working/tuesAMsourcelists/
 
 
*fall back on the [[Units]] page to look for a first pass at explanations and steps (see especially section 3.3)
 
*fall back on the [[Units]] page to look for a first pass at explanations and steps (see especially section 3.3)
 
*rely on the [[Central wavelengths and zero points]] page to look up wavelengths and zero points
 
*rely on the [[Central wavelengths and zero points]] page to look up wavelengths and zero points
Line 27: Line 33:
 
*when you are ready to interpret the SEDs (either as you go along or all at once after making these), you want to look for things that look like the 'template' SEDs (see [[http://coolwiki.ipac.caltech.edu/index.php/Studying_Young_Stars#More_in_depth.2C_and_SEDs here]] or my talk from july). things that don't look like the template, or things that don't have enough data to decide if they look like the template, or things that are missing bands.. those are the ones you want to note as "junk" or "check" or "wth?" etc.
 
*when you are ready to interpret the SEDs (either as you go along or all at once after making these), you want to look for things that look like the 'template' SEDs (see [[http://coolwiki.ipac.caltech.edu/index.php/Studying_Young_Stars#More_in_depth.2C_and_SEDs here]] or my talk from july). things that don't look like the template, or things that don't have enough data to decide if they look like the template, or things that are missing bands.. those are the ones you want to note as "junk" or "check" or "wth?" etc.
 
*when you set about interpreting these, you want to incorporate the knowledge we have already accumulated to this point. why are they in the short list of things we care about? (separate from "why did luisa put them in the list for us to do?" :) what i mean is "are they previously identified objects and/or are they objects xavier's color selection picked out?") what did we already note about these objects in the images?
 
*when you set about interpreting these, you want to incorporate the knowledge we have already accumulated to this point. why are they in the short list of things we care about? (separate from "why did luisa put them in the list for us to do?" :) what i mean is "are they previously identified objects and/or are they objects xavier's color selection picked out?") what did we already note about these objects in the images?
 +
*be careful about limits vs. detections!
  
 
ok, so, please try making SEDs for the objects with the following internal numbers:
 
ok, so, please try making SEDs for the objects with the following internal numbers:
Line 42: Line 49:
  
 
hopefully this will not be TOO painful. . . :)
 
hopefully this will not be TOO painful. . . :)
 +
 +
'''Clearer(?) version of instructions:''' for these 11 sources, there are TWO things to do:
 +
#make the SED. compare what you get to what i got, e.g., here: http://web.ipac.caltech.edu/staff/rebull/working/pages/ if you click on the page with the numbers corresponding to any of the numbers above, then you can see my SED. did i do it right? did you get the same thing?
 +
#then, once you have the SEDs (either as you go along, or once you have the whole set), set about interpreting them. see if you can look at them with a critical eye. which would you keep or drop? why?
 +
 +
=SED evaluation task - DUE 10/4/12=
 +
 +
(for the actual files, which i don't want to post on the open internet, see Email from me from 14 Sep)
 +
 +
(1) notesandnumbers20120914.xlsx - to create this, I did:
 +
* snapshot export of googledoc on tuesday
 +
* added some columns based on the master db plus other information, including what i did and why
 +
* one column is "final verdict" meaning "final image verdict"
 +
* one column has a yellow "see notes" box if the notes are particularly interesting.
 +
I provide this so that you have a complete record of what (you said) you did and what i did, and why. You might be interested in reading these sequentially, or looking for ones where your name is tagged, or any of a dozen other ways to manipulate these data. Please feel free to do so, but do NOT invest hours in trying to correct or backfill the assessments you're responsible for, because everyone made mistakes, I corrected them where I could, and where I couldn't, well, in any case, we've just moved forward from here.
 +
 +
(2) the outcome of this process was the 301 sources we will consider for SED assessment purposes. I have copied the rows from the first spreadsheet above into a new spreadsheet, deleted some now irrelevant columns, and added a new column for SED assessment.
 +
 +
AFTER you get done with the 'training' set of 11 SEDs, then the next step will be doing this SED assessment. All 301 have been dropped into http://web.ipac.caltech.edu/staff/rebull/working/pages/
 +
with the web page setup as before. Note that if a source doesn't appear in this shortlist of 301, it doesn't have a web page.
 +
 +
My questions to you:
 +
#There are 301 sources. Should every team look at every one, e.g., do you want to be parcelled out a subset and do just some, or should everyone do all of them? Looking at each one should go a LOT faster than the finder chart or wise image inspection. Thoughts?  '''YES, EVERYONE SHOULD LOOK AT ALL OF THEM.'''
 +
# i think the google doc failed in a very specific way -- people DID overwrite cells they weren't supposed to overwrite. we don't want to lose information. so, my inclination is thus to suggest taht you fill out column P in the attached with, well, "your SED assessment" and then mail it back to me when you're done, regardless of the answer to question 1 above. I will then consolidate them back again into one big xls document.
 +
 +
In any case, my *advice* to you:
 +
as per our discussion on the phone yesterday, one very specific failure mode was related to the row number being not equal to the object number. This time (a) there is a new column N which is a copy of the object number, (b) the column A object number is now frozen so it doesn't move, (c) the numbers are not sequential -- this xls only contains those objects having survived to this point. so, yes, the numbers have NOT been reassigned, they are the same object numbers as before, so no matter what happens with question 1 above, you will never be assigned "numbers 1-77" because the numbers of the surviving things are not in sequential 1-2-3 order. This means that they DO NOT correspond to the row numbers, right off the bat.  at the moment they are in fact sorted in numerical order, so the object number in row 3 is greater than that in row 2, etc.  You can sort it however you want. but right now ROW 2 is OBJECT 10, so even from the very beginning, hopefully that specific problem is ameliorated.
 +
 +
Those of you who have already done the 11 training SEDs and feel like they want to dive in should feel free to do so, since now you have the updated information that came as a result of the image inspection.
 +
 +
=Spitzer photometry task -- DUE 10/25/12???=
 +
 +
Remember that we have serendipitous Spitzer data all over this field. I could have sworn that I made this figure for you before, but it does not seem to be on the wiki, so here it is again. Background is WISE-1; polygons are the footprints of all the spitzer observations in this region.
 +
 +
[[image:spitzerfootprintsonw1_brc27.png]]
 +
 +
out of the nominally 'interesting' subsample (e.g., those tagged as previously known,
 +
plus those tagged by xavier as possibly YSOs, prior to any weeding based on images), I
 +
get 142 objects that have measurements in at least 1 band.
 +
 +
i have generated a table that has :
 +
*internal number
 +
*ra and dec (for checking!)
 +
*the list of IRAC AORKEYs
 +
*and the list of MIPS AORKEYs.
 +
 +
Recall that the AORKEY is the unique large integer number used to identify the observation in the mission. I have prepended a string that makes sense to me, so here is the decoding:
 +
*IRAC AORs
 +
**original17512192 - the original BRC 27 observation, AORKEY=17512192
 +
**brc2617511680 - the BRC 26 observation just to the south of BRC 27; AORKEY=17511680 (caution as the "26" is actually part of the prepended string)
 +
**steve14773760 = a cryo observation studying galactic structure, obtained by principal investigator steve majewski (hence 'steve'), AORKEY=14773760
 +
**glimpse38949120 = a warm IRAC (so only i1i2 not i3i4) observation from the GLIMPSE program, which studies galactic structure, AORKEY=38949120
 +
**glimpse38973184 = another GLIMPSE stripe, AORKEY=38973184
 +
**glimpse39074304 = another GLIMPSE stripe, AORKEY=39074304
 +
*MIPS AORs
 +
**There are two AORs, 17512448 and 17511936. both of them have data obtained on the target as specified ("prime") and data obtained serendipitously while the 70 um array was looking at the target as specified; these 24 um data are "nonprime". There are 4 possible mosaics you can use from these two AORs (two each, prime and non-prime). The notes file tells you which AORKEY and which mosaic (prime or non-prime).
 +
 +
Some of these sources are measured in >1 mosaic. Formally, we should measure in all of them and then take a weighted average of these measurements. some of these are listed as being in the IRAC AOR xxx but in reality the object falls in only 1 or 2 of the channels, depending on where the object and serendipitous data happen to be.
 +
 +
You can download AORs by AORKEY in the SHA.
 +
 +
Some objects might come and go out of this list as we sort out the issues with the image inspection.
 +
 +
=Timelines=
 +
 +
Working backwards...
 +
 +
*Let's print by Dec 14. So, the text needs to be done by then.
 +
*We ought to be passing around the final thing for comments by Dec 7.
 +
*The writing ought to be started by Thanksgiving weekend, Nov 22.
 +
*All the final weeding needs to be done by Nov 15 so that we have a clean sample to write about.
 +
*So, all the final photometry from Spitzer needs to be done by Oct 25 or Nov 1 so that we can chase irregularities, recheck images, remake SEDs and weed.
 +
*So all the SED assessments need to be done ASAP

Latest revision as of 23:54, 4 October 2012

Sep 6, 2012 - big picture and image assessments

File:Email20120906.txt - email from luisa with tasks and timeline questions from here on out

Notes from telecon:

  • if you can get your image assessements into the google doc in the next few days, i'll look at them all on monday and see if we can drop anything obviously bad.
  • i will pick a good set of seds tonight - 10 objects for everyone to make to show that they can do it, they can understand the concepts involved. those 10 will be picked to start training your brain on 'good' and 'bad' seds. those seds due next thursday.
  • bob will lead the edu poster; lauren will lead the sci poster. first drafts of those abstracts also due next thursday.
  • we will start looking at my seds and weeding based on those seds for the subset that survive the image weeding process starting next thursday, hopefully to be done in a week or at most 2.
  • not sure what to do re: optical data. do we have access to new data from HI (any of the telescopes)? or should we try PTF? or just go with what we have from last year as a place to start?
  • in any case, we need to get moving on the photometry on the spitzer images in the next 2-3 weeks.

Image Assessment task - given July, due Sep 10

There seems to have been a lot of confusion about copying stuff back into the master spreadsheet. File:Email20120910.txt - email from luisa

Email from Sep 14 included a 'final' file merging everyone's information and providing a 'final' verdict on every inspected item. Do NOT invest hours in trying to correct or backfill the assessments you're responsible for, because everyone made mistakes, I corrected them where I could, and where I couldn't, well, in any case, we've just moved forward from here. Outcome of this process was 301 sources, to be passed to SED assessment task, below.


SED training task - given 9/6/12, due 9/13/12 9/20/12

here is the list of SEDs to make more or less by hand.

i tried to pick a range of shapes, range of kinds of data, range of new/known, etc. some of these will have optical data (of two different kinds!), as well as NIR (2mass) data, WISE, and sometimes Spitzer data. i recommend doing them in this order, or at least trying to interpret them in this order, because i buried some tricky ones in here.

guidelines:

  • start with your notes from july and the full catalog i gave you (at least i think i gave you) that has ALL the bands currently in my master DB, optical through 70 um. if you can't find it, it's still here: http://web.ipac.caltech.edu/staff/rebull/working/tuesAMsourcelists/
  • fall back on the Units page to look for a first pass at explanations and steps (see especially section 3.3)
  • rely on the Central wavelengths and zero points page to look up wavelengths and zero points
  • don't forget that AB magnitudes are different than Vega magnitudes.
  • email questions when you have them; we can't necessarily save them all for next thursday if they are showstopping issues!
  • when you are ready to interpret the SEDs (either as you go along or all at once after making these), you want to look for things that look like the 'template' SEDs (see [here] or my talk from july). things that don't look like the template, or things that don't have enough data to decide if they look like the template, or things that are missing bands.. those are the ones you want to note as "junk" or "check" or "wth?" etc.
  • when you set about interpreting these, you want to incorporate the knowledge we have already accumulated to this point. why are they in the short list of things we care about? (separate from "why did luisa put them in the list for us to do?" :) what i mean is "are they previously identified objects and/or are they objects xavier's color selection picked out?") what did we already note about these objects in the images?
  • be careful about limits vs. detections!

ok, so, please try making SEDs for the objects with the following internal numbers:

  • 373
  • 1364
  • 1410
  • 1491
  • 1563
  • 483
  • 826
  • 1547
  • 1399
  • 1261
  • 1535

hopefully this will not be TOO painful. . . :)

Clearer(?) version of instructions: for these 11 sources, there are TWO things to do:

  1. make the SED. compare what you get to what i got, e.g., here: http://web.ipac.caltech.edu/staff/rebull/working/pages/ if you click on the page with the numbers corresponding to any of the numbers above, then you can see my SED. did i do it right? did you get the same thing?
  2. then, once you have the SEDs (either as you go along, or once you have the whole set), set about interpreting them. see if you can look at them with a critical eye. which would you keep or drop? why?

SED evaluation task - DUE 10/4/12

(for the actual files, which i don't want to post on the open internet, see Email from me from 14 Sep)

(1) notesandnumbers20120914.xlsx - to create this, I did:

  • snapshot export of googledoc on tuesday
  • added some columns based on the master db plus other information, including what i did and why
  • one column is "final verdict" meaning "final image verdict"
  • one column has a yellow "see notes" box if the notes are particularly interesting.

I provide this so that you have a complete record of what (you said) you did and what i did, and why. You might be interested in reading these sequentially, or looking for ones where your name is tagged, or any of a dozen other ways to manipulate these data. Please feel free to do so, but do NOT invest hours in trying to correct or backfill the assessments you're responsible for, because everyone made mistakes, I corrected them where I could, and where I couldn't, well, in any case, we've just moved forward from here.

(2) the outcome of this process was the 301 sources we will consider for SED assessment purposes. I have copied the rows from the first spreadsheet above into a new spreadsheet, deleted some now irrelevant columns, and added a new column for SED assessment.

AFTER you get done with the 'training' set of 11 SEDs, then the next step will be doing this SED assessment. All 301 have been dropped into http://web.ipac.caltech.edu/staff/rebull/working/pages/ with the web page setup as before. Note that if a source doesn't appear in this shortlist of 301, it doesn't have a web page.

My questions to you:

  1. There are 301 sources. Should every team look at every one, e.g., do you want to be parcelled out a subset and do just some, or should everyone do all of them? Looking at each one should go a LOT faster than the finder chart or wise image inspection. Thoughts? YES, EVERYONE SHOULD LOOK AT ALL OF THEM.
  2. i think the google doc failed in a very specific way -- people DID overwrite cells they weren't supposed to overwrite. we don't want to lose information. so, my inclination is thus to suggest taht you fill out column P in the attached with, well, "your SED assessment" and then mail it back to me when you're done, regardless of the answer to question 1 above. I will then consolidate them back again into one big xls document.

In any case, my *advice* to you: as per our discussion on the phone yesterday, one very specific failure mode was related to the row number being not equal to the object number. This time (a) there is a new column N which is a copy of the object number, (b) the column A object number is now frozen so it doesn't move, (c) the numbers are not sequential -- this xls only contains those objects having survived to this point. so, yes, the numbers have NOT been reassigned, they are the same object numbers as before, so no matter what happens with question 1 above, you will never be assigned "numbers 1-77" because the numbers of the surviving things are not in sequential 1-2-3 order. This means that they DO NOT correspond to the row numbers, right off the bat. at the moment they are in fact sorted in numerical order, so the object number in row 3 is greater than that in row 2, etc. You can sort it however you want. but right now ROW 2 is OBJECT 10, so even from the very beginning, hopefully that specific problem is ameliorated.

Those of you who have already done the 11 training SEDs and feel like they want to dive in should feel free to do so, since now you have the updated information that came as a result of the image inspection.

Spitzer photometry task -- DUE 10/25/12???

Remember that we have serendipitous Spitzer data all over this field. I could have sworn that I made this figure for you before, but it does not seem to be on the wiki, so here it is again. Background is WISE-1; polygons are the footprints of all the spitzer observations in this region.

Spitzerfootprintsonw1 brc27.png

out of the nominally 'interesting' subsample (e.g., those tagged as previously known, plus those tagged by xavier as possibly YSOs, prior to any weeding based on images), I get 142 objects that have measurements in at least 1 band.

i have generated a table that has :

  • internal number
  • ra and dec (for checking!)
  • the list of IRAC AORKEYs
  • and the list of MIPS AORKEYs.

Recall that the AORKEY is the unique large integer number used to identify the observation in the mission. I have prepended a string that makes sense to me, so here is the decoding:

  • IRAC AORs
    • original17512192 - the original BRC 27 observation, AORKEY=17512192
    • brc2617511680 - the BRC 26 observation just to the south of BRC 27; AORKEY=17511680 (caution as the "26" is actually part of the prepended string)
    • steve14773760 = a cryo observation studying galactic structure, obtained by principal investigator steve majewski (hence 'steve'), AORKEY=14773760
    • glimpse38949120 = a warm IRAC (so only i1i2 not i3i4) observation from the GLIMPSE program, which studies galactic structure, AORKEY=38949120
    • glimpse38973184 = another GLIMPSE stripe, AORKEY=38973184
    • glimpse39074304 = another GLIMPSE stripe, AORKEY=39074304
  • MIPS AORs
    • There are two AORs, 17512448 and 17511936. both of them have data obtained on the target as specified ("prime") and data obtained serendipitously while the 70 um array was looking at the target as specified; these 24 um data are "nonprime". There are 4 possible mosaics you can use from these two AORs (two each, prime and non-prime). The notes file tells you which AORKEY and which mosaic (prime or non-prime).

Some of these sources are measured in >1 mosaic. Formally, we should measure in all of them and then take a weighted average of these measurements. some of these are listed as being in the IRAC AOR xxx but in reality the object falls in only 1 or 2 of the channels, depending on where the object and serendipitous data happen to be.

You can download AORs by AORKEY in the SHA.

Some objects might come and go out of this list as we sort out the issues with the image inspection.

Timelines

Working backwards...

  • Let's print by Dec 14. So, the text needs to be done by then.
  • We ought to be passing around the final thing for comments by Dec 7.
  • The writing ought to be started by Thanksgiving weekend, Nov 22.
  • All the final weeding needs to be done by Nov 15 so that we have a clean sample to write about.
  • So, all the final photometry from Spitzer needs to be done by Oct 25 or Nov 1 so that we can chase irregularities, recheck images, remake SEDs and weed.
  • So all the SED assessments need to be done ASAP