What did luisa do?

From CoolWiki
Revision as of 15:54, 9 December 2008 by Rebull (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

I made this into a separate page so as not to ruin your discussion among your classes. :)

in the interest of getting information into the page, i am using minimal punctuation and grammar and am being perhaps too brief in some of these steps. let me know when something doesn't make sense.

Left as an exercise for the reader: Compare these steps to those we discussed in June in the Working with L1688 analysis. hint: those steps that seem too briefly summarized HERE have a LOT more explanation on that other page.


LDN 425

NB: discussed in june.

download bcd and pbcd data. likely only to use pbcd, but might as well get bcd now in case i have to reprocess it.

unzip it.

look at pipeline produced mosaics. they look fine. no need to reprocess.

make some 3-color mosaics. note bright red thing in depths of the stuff that is dark in the optical. no other screamingly obvious objects seen in 3-color mosaic as bright red things. another obj seen at 24, but not as red as the first obj.

look in literature for previously known objects. note that the red thing is the same object studied by connelley et al. (2007 AJ 133, 1528).

do photometry, a.k.a. run source extraction (manually on previously known obj, or automatically on everything.) i'm using apex and idl, but APT is also JUST FINE. check source extraction to make sure photometry makes sense and looks good. it does.

no additional very red sources found in data. all other sources detected at multiple bands seem to have zero spitzer color, e.g., they are not red. not all of these sources are necessarily stars, but if they are stars, then they don't have any infrared excess and thus we infer no circumstellar dust. HOWEVER, the one visibly red object is quite impressive and very red indeed.

several people measured and reported fluxes at various bands. note that PA used MaxImDL, Rebull used apex/idl, and the rest used APT:

person or group irac1 (Jy) irac2 (Jy) irac3 (Jy) irac4 (Jy) mips1 (Jy)
PA group (Spuck et al.) 0.0184 0.0406 0.0634 0.0771 1.58
DeWolf & Johnson team 0.0173 0.0394 0.06194 0.0793 1.822
Shaefers 0.01783 0.03963 0.06626 0.0916 1.9068
NY group (Guastella et al.) 0.017106 0.039 0.0625 0.0779 1.82
MT Group (Dave McDonald) 0.017193
MT group (Brandi, Blair, and Jacob average) 0.0172 0.0389 0.0617 0.0778 1.636
Rebull 0.0152 0.0349 0.0600 0.0721 1.363
AVERAGE 0.0172 0.0388 0.0625 0.0788 1.675
STD DEV 0.000814 0.001665 0.001837 0.005570 0.172084
fractional error 0.047 0.042 0.029 0.070 0.102

the last row means that there is a 5% scatter in our measurements for irac-1 and 2, a 3% scatter in the measurements for irac-3, a 7% error in the measurements for irac-4, and a 10% error in the measurements for mips-1. these are really pretty darn good! pat yourself on the back. there is rather more scatter in mips-1 in particular than i was expecting, but this is probably because APT doesn't allow you to center up on a sub-pixel value.

when we measure spitzer fluxes, we get them back in uJy (or mJy or similar). the 2mass data come to us in magnitudes. we can use both magnitudes and fluxes to assess how red this source is. it is useful thus to convert the measured spitzer fluxes to mags and the 2mass mags to fluxes. in june, we wrote a spreadsheet to do that. let me know if you need a copy, or write one from scratch using the stuff on the L1688 page (or, more specifically, pages it references).

go look at the CMDs and CCDs that people use to find YSOs. i gave you several in the intro ppt, several are buried in the wiki, and you can find many more in the literature. Calculate the colors for our red object and see where it falls in these diagrams. is this a weak or strong excess source? the spreadsheet we made in june calculates several colors for you.

go look at the SEDs that are expected for YSOs. i gave you several in the intro ppt, several are buried in the wiki, and you can find many more in the literature. plot up the SED for this source. watch your units (as it says elsewhere in the wiki) because this can be VERY tricky. you will find that you are either right or very very wrong (e.g., off by 38 orders of magnitude, etc). see how the SED for this source compares to other YSOs. is this a weak or strong excess source? the spreadsheet we made in june makes the SED.

taking a shot in the dark, i wrote connelley to see if we could get copies of his fits images from his paper. he obliged and sent them to us, so we now have very deep K-band imaging, much deeper than that from 2mass. but those data don't have any WCS information in the headers, so they are likely to be hard to use in MaxImDL.

the SED and a table of the colors can go into the poster. a CMD or CCD with a single point is unlikely to be helpful, but if you wish to put it in context, you can add your point to one of the existing plots you find in the literature or on the wiki and put THAT on the poster. make sure you cite where the original plot came from (and specifically what region the rest of the data points come from).

good enough now to stop here for the AAS poster discussion on this target. NOTE THAT if you want any of the plots i made for the poster, i can make them using bigger font. just let me know.

ultimately, we would like to model this source, looking at the shape of the SED AND (this is important) the morphology of the object as it appears in the deep K image and the irac-1 image. it is interesting that is is not a point but more of a rectangle. this shape plus the SED is telling us something about the nature of the source. russ's spreadsheet will allow us to model it as a series of blackbody curves, but i'm talking about something more sophisticated (see links on l1688 page for a start).

LDN 981

download bcd and pbcd data. likely only to use pbcd, but might as well get bcd now in case i have to reprocess it.

unzip it.

look at pipeline produced mosaics. they look fine. no need to reprocess.

make some multi-color images. look for very red sources in any combination of images. there aren't any boomers like in the other cloud, but i think i see some red objects anyway.

do photometry, a.k.a. run source extraction (i'm using apex and for irac, idl as well, but APT is also JUST FINE). check source extraction to make sure photometry makes sense and looks good. it does. TONS more sources here than in the other observation, so it is easier to make this assessment. Luisa's plots of lynds981 photometry (pdf) now that there are a lot more points, the things that are red kinda stand out in these plots (the analogous plots for our other data look really silly because there are so few points).

UPDATE to this, 14:59, 9 December 2008 (PST): in looking to see if i found some of the objects that show up as visibly red in Jennifer's composite image, i discovered that i had only found 2 of the objects successfully, and a third is too faint/diffuse for the source extractor to find. but the remaining two of them are missing from the catalog for reasons i initially didn't understand. in BOTH cases, in one band (not the same one in both cases), the source extractor saw two sources where in reality there is only one. the other bands then automatically got matched, half to one of the sources, the other half to the other source, resulting in an incomplete set of measurements and thus the source didn't get selected in the color-mag diagrams. i didn't realize that the source detection algorithm as run on this data had this problem. it's had that problem on some other data that i have, but i didn't think the source density was high enough here that that would matter, but it does. this kind of problem would not have been revealed by the plot above; it's only going to show up by hand-inspection of individual sources (which is what i did when i went to see about finding matches to the visibly red sources).

we now return to the previous narrative:

go look at the CMDs and CCDs that people use to find YSOs. i gave you several in the intro ppt, several are buried in the wiki, and you can find many more in the literature. look at where the objects are here that fall in the "red zone" and see how they compare to YSOs in those other plots.

here are the color selections i have most recently used in my work elsewhere to find YSOs. note that i am looking for not just red but also the brighter things (note, "brighter" does NOT mean saturated!! one of the comments on the talk page seems to imply some of you are looking for saturated objects). "Ks" = "K-short", a.k.a. the 2MASS K band measurement. the equations describe lines in the color-mag and color-color plots (try 'em! plot 'em up and see!). the color-mag and color-color spaces i used are:

  • Ks vs. Ks-[24]
  • [8] vs. [8]-[24]
  • [4.5] vs. [4.5]-[8]
  • [3.6]-[4.5] vs. [5.8]-[8] with an additional [3.6] brightness cutoff

the formulas are here: Ccdselection.png

this color-based selection of YSOs suggests the following candidates (NOTE THAT you should refer to these by their full coordinates -- the names that are here follow IAU convention for new objects):

  *****OLD LIST:*****
| name             |RA         |Dec        |
|                  |d          |d          |
 205939.52+502124.2  314.914683   50.356735
 205947.91+501432.3  314.949627   50.242329
 210017.34+501940.4  315.072276   50.327901
 210017.36+501940.3  315.072334   50.327888
 210036.99+502057.6  315.154161   50.349361
 210046.34+502344.9  315.193086   50.395817
 210049.05+501524.5  315.204405   50.256821
 210049.50+501542.9  315.206282   50.261929
 210054.66+502617.0  315.227775   50.438076
 210120.34+502007.7  315.334766   50.335499
 210142.11+501305.4  315.425497   50.218189
 210147.48+501821.9  315.447873   50.306110
  *** NEW LIST **** some obj (duplicates) went away, some new
         name        |RA       |Dec        |
                     |d        |d          |
 205939.52+502124.2  314.914683   50.356735 
 205947.91+501432.3  314.949627   50.242329 
 210017.36+501940.3  315.072334   50.327888 
 210036.99+502057.6  315.154161   50.349361 
 210046.34+502344.9  315.193086   50.395817 
 210049.05+501524.5  315.204405   50.256821 
 210049.50+501542.9  315.206282   50.261929 
 210052.57+501254.3  315.219062   50.215104 
 210054.66+502617.0  315.227775   50.438076 
 210109.38+501048.1  315.289102   50.180035 
 210120.34+502007.7  315.334766   50.335499 
 210142.11+501305.4  315.425497   50.218189 
 210147.48+501821.9  315.447873   50.306110 

a few of these are found in more than 1 CMD, suggesting perhaps more likely YSO? here are some sample CMDs/CCDs with these objects highlighted in red - you can see they are distinctly redder than the rest of the sample: Media:lynds981ysocandcmds.pdf UPDATED

also need to look at where in space they fall OLD IMAGE: Wherenewobj.png NEW IMAGE: Wherenewobj2.png i don't believe the photometry for those guys on the edge of the frame. (this is an irac-4 image, BTW.) i also question the two objects that are right on top of each other - i think there is probably just one source really there that the source extractor saw as two objects. (NOTE: this should have been a tip-off to me that there was something wrong! but anyway...)

UPDATE to this, 14:59, 9 December 2008 (PST): note that the objects that appear visibly red in the composite that Jennifer made are faint ... just too faint to have been caught by the specific YSO net i cast above. That's why they are still not on the list. They could still be legitimate YSOs, but I am worried because they are faint. If you want my photometry for these objects for comparison, my numbers are:

 210006.49+501214.7:
  IRAC1MAG  15.37
  IRAC2MAG  14.83
  IRAC3MAG  14.23
  IRAC4MAG  12.12
  MIPS1MAG  8.93
 210104.02+501748.1   
  IRAC1MAG  14.26
  IRAC2MAG  13.70
  IRAC3MAG  12.88
  IRAC4MAG  11.71
  MIPS1MAG  7.13


we now return you to the previous narrative.

are any of these previuosly known objects? go get literature. quanz et al. (2007, ApJ, 656, 287, also astro-ph/0610786) paper leads us to feigelson and kriss (1983, AJ, 88, 431) paper. feigelson and kriss report only names and finder charts; need to go to 2mass to extract high-quality modern coordinates. use their finder chart, a new poss image, and a new 2mass image to reconstruct which obj are which and make note of coords and JHK mags. can find all but one based on finder chart, comparing images. got new 2mass coords for all of those:

 \char comment = cleaned-up feigelson & kriss catalog 
 | name           |RA         |Dec        |  
 |                |d          |d          |  
               1  315.511000   50.495400
               2  315.473400   50.537900
               3  315.362247   50.528336
               4  315.304731   50.401846
               5  315.287700   50.361600
               6  315.244400   50.403600
               7  315.184000   50.488800
               8  315.172215   50.495880
               9  315.155700   50.500000
              10  315.134800   50.527100
              12  315.022700   50.437300
              13  315.070701   50.401503
              14  315.136002   50.363911
              15  315.041164   50.352451
              16  315.175594   50.259917
              17  315.182015   50.204971
              18  315.221231   50.271915
              19  315.285914   50.260887
              20  315.355610   50.330031
              21  315.344357   50.360091
              22  315.536100   50.378600


find 9 with counterparts in our images, none of which are disk candidates based on above ("fk" numbers correspond to numbers above):

fk15  21 00 09.9  +50 21 09
fk13  21 00 17.0  +50 24 05
fk14  21 00 32.6  +50 21 50
fk 8  21 00 41.3  +50 29 45
fk17  21 00 43.7  +50 12 18
fk18  21 00 53.1  +50 16 19
fk19  21 01 08.6  +50 15 39
fk 4  21 01 13.2  +50 24 07
fk21  21 01 22.7  +50 21 36

here is all my photometry on the red things, a.k.a. yso candidates: Media:ldn981ysocand.cat.tbl (UPDATED!). if you want my photometry for other objects (like any of the ones you all found on your own), let me know; i just didn't post the whole catalog because it would be too overwhelming.

AS ABOVE, when we measure spitzer fluxes, we get them back in uJy (or mJy or similar). the 2mass data come to us in magnitudes. it is useful thus to convert the measured spitzer fluxes to mags and the 2mass mags to fluxes. in june, we wrote a spreadsheet to do that. i did that in the file above.

go look at the CMDs and CCDs that people use to find YSOs. look above at where our objects fall in these diagrams. are they weak or strong compared to the source in our other cloud?

go look at the SEDs that are expected for YSOs. i gave you several in the intro ppt, several are buried in the wiki, and you can find many more in the literature. plot up the SED for these sources. as above, watch your units. are these sources weak or strong compared to the source in our other cloud? here are my SEDs: Media:ldn981ysocandseds.pdf (UPDATED) most of them are pretty small excess objects. this is consistent with what our eyes told us way back at the beginning - there aren't any booming red objects here.

here, one or two CMDs and/or SEDs can go into the poster.

good enough now to stop here for the AAS poster discussion on this target. NOTE THAT if you want any of the plots i made for the poster, i can make them using bigger font. just let me know.

ultimately, we want to go get follow-up optical classification spectra for these things to verify that they are really YSOs and not just background objects (or errors in the photometry).

oh, and we should look into how much of the SED we can get for V1331 Cyg (or whatever that thing is that is named something like that, with the ring around it in the DSS image).

Personal tools