Source Classification v2

Keywords: All you can find under object_type column

Zero
2mass
one, red1
two, red2
fd_L0V, fd_L5V, fd_T0V, fd_T5V
star, star_{spectral types}, star_no_BANDx      [BANDx=J, H, Ks, IR1, IR2, IR3, IR4, or MP1]
YSOc_IR4_PAH-em, YSOc_MP1_PAH-em, YSOc_IR4+MP1_PAH-em
YSOc_IR4_star+dust(BANDx), YSOc_MP1_star+dust(BANDx), YSOc_IR4+MP1_star+dust(BNADx)    [BANDx=IR1, IR2, IR3, IR4, or MP1]
YSOc_IR4_red, YSOc_MP1_red, YSOc_IR4+MP1_red
YSOc_IR4, YSOc_MP1, YSOc_IR4+MP1
PAH-em,
star+dust(BANDx)    [BANDx=IR1, IR2, IR3, IR4, or MP1]
red
rising
falling
cup-up
cup-down
flat

Conditions for each catagories:

Zero (Attention! asteroid lovers)
Sources not detected in combined epochs, but has at least one band detection in one of the epochs
2mass

We propose not dealing with "2MASS only" sources. Most of them are in MIPS only regions and are likely to be stars.

one, red1
Sources has only one band detection. red1 is detected in IR4 OR MP1 band.

two, red2
Sources has two band detections. red2 is detected in IR4 AND MP1.

fd_{spectral types}
brown dwarf cadidates, selected with SED modeling fitting, using the observed field dwarf colors from Patten et al.

star, star_{spectral types}, star_no_BANDx
spectral types ranges from B0I to M5V, BANDx ranges from J to MP1
(Note: if BANDx is the last band, the source could be a star+disk candidate)
PAH-em:
[3.6]-[4.5]<0.6, 1.5<[5.8]-[8.0] ( equivalent to F(I1)/F(I2) > 0.9, F(I4)/F(I3) > 2.2 )

Star+dust
Star criteria: at least 3 bands fit stellar SED models
Disk criteria: triggered when one band flux excess > 3 sigma, then bands with longer wavelengths should have flux excess > 2 sigma
(Old disk criteria: IR4 or MP1 excess > 3 sigma)
including star(J-Ks)+disk, star(J-IR1)+disk, ..., star(J-IR4)+disk

Red:
MIPS1 flux > 3 x nearest IRAC flux (sources have at leasst 3 band detections)
"YSOc_IR4" or "YSOc_MP1" or "YSOc_IR4+MP1" for YSO candidates

"YSOc_IR4"

[4.6]-[8.0] > 0.5 [ equivalent to F(IR2)/F(IR4) < 1.8 or F(IR4)/F(IR2) > 0.56] : Separate star component
[8.0] < 14- ([4.6]-[8.0]) [ equivalent to F(IR2) > 0.45 mJy ] : excluding ~95% of SWIRE sources

"YSOc_MP1"

[8]-[24] >0.7 [ equivalent to F(IR4)/F(MP1) < 4.6 or F(MP1)/F(IR4) > 0.22] : Separate star component

[24] < 12-([8]-[24]) [ equivalent to F(IR4) > 1 mJy ] : excluding ~95% of SWIRE sources

SED morphologies
First divide SED into 3 piece (left, center, right), then use 2 sigma criteria to compare the average of each piece
if nbands=3,4,5    left and right each has one band, center has 1 or 2 or 3 bands
   nbands=6,7 left and right each has two bands, center has 2 or 3 bands
   nbands=8    left has 3 bands (J, H, Ks), right has 2 bands (IR4 and MP1), center has 3 bands (IR1 to IR3)

rising left-center < -2 sigma AND right - center > 2 sigma

falling left-center > 2 sigma AND right - center < - 2 sigma

cup-up ( left + right ) / 2 - center > 2 sigma

cup-down ( left + right ) / 2 - center < -2 sigma

flat -2 sigma < ( ( left + right )/2 - center ) < 2

8/31/05

Criteria for selecting YSO candidates

First set:

[4.6]-[8.0] > 0.5 [ equivalent to F(IR2)/F(IR4) < 1.8 or F(IR4)/F(IR2) > 0.56] : Separate star component
[8.0] < 14- ([4.6]-[8.0]) [ equivalent to F(IR2) > 0.45 mJy ] : excluding ~95% of SWIRE sources

Second set:

[8]-[24] >0.7 [ equivalent to F(IR4)/F(MP1) < 4.6 or F(MP1)/F(IR4) > 0.22] : Separate star component
[24] < 12-([8]-[24]) [ equivalent to F(IR4) > 1 mJy ] : excluding ~95% of SWIRE sources

(If these plots are too small for you, make your netscape window wider OR use "View Image" with right click)

Using all SWIRE sources in 7.5 square degree area with flux quality A or B for IRAC 1-4 bands and MP1 band

	total	2mass	1 band sources		2 band soruces		3+ band sources (percentages = numbers / total sources with at least 3 band detections) YSO candidates fit (IR2,IR4) criteria ADDITIONAL YSO candidates fit (IR4, MP1) criteria, but not (IR2,IR4) criteria non-YSO candidates									total non-star sources total YSO from (IR2,IR4) additional YSO from (IR4, MP1) total non-YSO
	total	2mass	red1	one	red2	two	star	PAH-em	star+dust	red	rising	falling	cup-up	cup-down	flat
SWIRE (~1 deg^2)	15410	56	23	443	2	6914	2989 (37.49%)	455 ( 5.71%) 33 36 386	784 ( 9.83%) 10 3 774	2483 (31.15%) 3 0 2480	334 ( 4.19%) 6 4 324	55 ( 0.69%) 0 0 55	541 ( 6.79%) 3 1 537	94 ( 1.18%) 1 1 92	237 ( 2.97%) 1 0 236	4983 57 45 4884
SWIRE with c2d limits	3205	58	0	2	0	131	2347 (77.87%)	275 ( 9.12%) 33 36 206	133 ( 4.41%) 10 1 122	56 ( 1.86%) 3 0 53	45 ( 0.30%) 6 4 35	48 ( 0.32%) 0 0 48	48 ( 0.32%) 3 1 44	47 ( 0.31%) 1 1 45	15 ( 0.10%) 1 0 14	667 57 43 567
Serpens OC	18122	14032	28	416	0	758	2793 (96.71%)	4 ( 0.14%) 0 1 3	26 ( 0.90%) 0 1 25	2 ( 0.07%) 0 0 2	2 ( 0.07%) 0 0 2	14 ( 0.48%) 0 0 14	34 ( 1.18%) 0 0 34	10 ( 0.35%) 0 0 10	3 ( 0.10%) 0 0 3	95 0 2 93
Serpens (~ 1 deg^2)	67702	12674	146	7974	4	14154	31311 (95.61%)	55 ( 0.17%) 9 15 31	159 ( 0.49%) 108 23 28	60 ( 0.18%) 27 2 31	83 ( 0.25%) 39 7 37	205 ( 0.30%) 0 1 204	518 ( 1.58%) 2 14 502	299 ( 0.91%) 70 4 225	60 ( 0.18%) 11 0 49	1439 266 66 1107

8/24/05

Compare SWIRE data and Serpens data

1 square deg of SWIRE data cut with c2d detection limits

Sources are classified following the sequence from left to right, each source fall in only one category

	total	2mass	1 band sources		2 band soruces		3+ band sources (percentages = numbers / total sources with at least 3 band detections)
	total	2mass	red1	one	red2	two	star	star+disk	red	PAH-em	rising	falling	cup-up	cup-down	flat
SWIRE (~1 deg^2)	15410	56	23	443	2	6914	3047 ( 38.22%)	1013 ( 12.71%)	2503 ( 31.40%)	148 ( 1.86%)	334 ( 4.19%)	55 ( 0.69%)	541 ( 6.79%)	94 ( 1.18%)	237 ( 2.97%)
SWIRE with c2d limits	3205	58	0	2	0	131	2372 ( 15.59%)	271 ( 1.78%)	66 ( 0.43%)	102 ( 0.67%)	45 ( 0.30%)	48 ( 0.32%)	48 ( 0.32%)	47 ( 0.31%)	15 ( 0.10%)
Serpens OC	18122	14032	28	416	0	758	2796 ( 96.81%)	26 ( 0.90%)	2 ( 0.07%)	1 ( 0.03%)	2 ( 0.07%)	14 ( 0.48%)	34 ( 1.18%)	10 ( 0.35%)	3 ( 0.10%)
Serpens (~ 1 deg^2)	67702	12674	146	7974	4	14154	31327 ( 95.65%)	180 ( 0.55%)	61 ( 0.19%)	17 ( 0.05%)	83 ( 0.25%)	205 ( 0.30%)	518 ( 1.58%)	299 ( 0.91%)	60 ( 0.18%)

star+disk, Red, PAH-em are competing with each other!

Source counts are different if we change the pick up sequence to PAH-em, star+disk, red

	PAH-em	star+disk	red
SWIRE	455 ( 5.71%)	784 ( 9.83%)	2483 ( 31.15%)
SWIRE with c2d limits	275 ( 1.81%)	133 ( 0.87%)	56 ( 0.37%)
Serpens OC	4 ( 0.14%)	26 ( 0.90%)	2 ( 0.07%)
Serpens	55 ( 0.17%)	159 ( 0.49%)	60 ( 0.18%)

Color-color diagrams

Plot Notations:
Points in these color-color diagrams show EVERY source that will be picked up by the reqired criteria. So, if you see one red and one green dot at the same location, that means the source will be pick up by both red and PAH-em criteria. Since we only assign ONE category for each source, the assigned category in the final cataloges will depend on the exact sequence we pick up sources. Note that these plots only show Star+disk, Red, PAH-em sources; they do not show rising, falling, cup-up, cup-down, and flat.

What do we see:

SWIRE is an almost pure galaxy sample -> almost all galaxies can be classified as star+disk sources.
SWIRE (toward galactic pole) and Serpens (on galactic plane) have very different source distribution on the (IR3-IR4, IR4-MP1) color-color diagram.
Not all galaxies can be pick up with PAH emission feature. However, since Serpens has very few PAH-em sources, PAH-em seem to be an adquent way to pick up a rather pure galaxy sample and [5.8]-[8.0] > 1.5 seem to be a proper value to pick up PAH-em sources.
Star+disk in Serpens mostly concentrate on one branch on the (IR3-IR4, IR4-MP1) color-colar diagram. Very few SWIRE sources has the same color as the sources in this star+disk branch.

Conclusion:
We suggest to pick up source with the sequence of PAH-em -> star+disk -> red.
star+disk goes before red, because it gives more constrains on the source properties.

Implications:

PAH-em are mostly galaxies. However, be aware that IF some true star+disks do have strong PAH emission feature, they will be classified as PAH-em, not star+disk.
Star+disk sources has little galaxy contamination if [8.0]-[24] < 3.5 and much larger galaxy contamination if [8.0]-[24] > 3.5.