star f ormation taste tests - university of...
TRANSCRIPT
Star Formation Taste TestsAlyssa A. Goodman
Harvard-Smithsonian Center for Astrophysics&
Initiative for Innovative Computing at Harvard
Taste Tests?“Taste Tests”? We frame this project by analogy. How does a great chef, making a complicated dish, know if she has created what she originally intended when she is done cooking? She “tastes.” She informs her cooking with her extensive knowledge of food chemistry (analytic theory), uses all the cooking equipment (simulations) she has in the kitchen to try to make something edible and tasty (starforming, and realistic), and then she uses her senses (observations) to see if what she made tastes as intended. “Tasting” in cooking actually encompasses the joint action of many senses: we propose here a combination of statistical techniques that we call “taste tests.” The tests will allow us to discerningly decide if what we sense (observe) and what we can cook (simulate) might actually be tasty (form stars), and how (analytic theory) that happens.
from: Goodman & Rosolowsky, NSF Proposal Fall 2006
Getting to “Observational Space”
Includes
–Radiative Transfer
–Projection to 2D sky plane, or “3D” of spectral-line data cubes
–Adding appropriate noise
– Imposing observing characteristics of a telesope
Example: The Spectral Correlation Function
(Padoan, Goodman & Juvela 2003)
COMPLETE Collaborators, Spring 2007:
Alyssa A. Goodman (CfA/IIC)
João Alves (Calar Alto, Spain)
Héctor Arce (AMNH)
Michelle Borkin (CfA/IIC)
Paola Caselli (Arcetri, Italy/CfA)
James DiFrancesco (HIA, Canada)
Jonathan Foster (CfA, PhD Student)
Sebastian Guillot (U. Victoria, Canada)
Mark Heyer (UMASS/FCRAO)
Doug Johnstone (HIA, Canada)
Jens Kauffmann (CfA/IIC)
Helen Kirk (HIA, Canada)
Di Li (JPL)
Jason Li (Harvard College)
Jaime Pineda (CfA, PhD Student)
Erik Rosolowsky (CfA)
Scott Schnee (Caltech)
Mario Tafalla (OAN, Spain)
COordinated Molecular Probe Line Extinction Thermal Emission Survey of Star-Forming Regions=
+
TasteTests
COMPLETE Collaborators, Spring 2007:
Alyssa A. Goodman (CfA/IIC)
João Alves (Calar Alto, Spain)
Héctor Arce (AMNH)
Michelle Borkin (CfA/IIC)
Paola Caselli (Arcetri, Italy/CfA)
James DiFrancesco (HIA, Canada)
Jonathan Foster (CfA, PhD Student)
Sebastian Guillot (U. Victoria, Canada)
Mark Heyer (UMASS/FCRAO)
Doug Johnstone (HIA, Canada)
Jens Kauffmann (CfA/IIC)
Helen Kirk (HIA, Canada)
Di Li (JPL)
Jason Li (Harvard College)
Jaime Pineda (CfA, PhD Student)
Erik Rosolowsky (CfA)
Scott Schnee (Caltech)
Mario Tafalla (OAN, Spain)
Meaningful structure in position-position-velocity space
Modeling l.o.s. temperature fluctuations
What stars form from what gas, when?
COordinated Molecular Probe Line Extinction Thermal Emission Survey of Star-Forming Regions=
Is the gas density distribution lognormal or not?
And scattering! (Cloudshine models)
Errors introduced by the assumption of isothermal dust along each line of sight
Variable fraction of emission from transiently heated very small dust grains
Variable dust properties (e.g. emissivity or emissivity spectral index)
1.2
1.0
0.8
0.6
0.4
0.2
0.0
log
(NIR
AS [m
ag A
V])
1.21.00.80.60.40.20.0
log (N2MASS [mag AV])
1.2
1.0
0.8
0.6
0.4
0.2
0.0
greyscale shows log (N
13CO [m
ag AV ])
2MASS/NICER
IRIS
Schnee, Bethell & Goodman 2006
Modeling line-of-sight temperature fluctuations
MHD Simulation+Radiative Xfer Code (No NOISE)
100/240 micron
Schnee, Bethell & Goodman 2006
60/100 micron
True AV
A V F
rom
Em
issi
on
A V F
rom
Em
issi
on
0.1
0.10.1
0.11.0 1.0 10.010.0
1.0
10.0
1.0
10.0
…and the correlationgets tighter still
at longer λ’s
True AV
Modeling line-of-sight temperature fluctuations
Tasting the Simulations
AV from NIR AV from NIR
60/100
True AV
Schnee, Bethell & Goodman 2006
A V F
rom
Em
issi
onModeling line-of-sight temperature fluctuations
I’ll show you later why that’s “True Av”
2006
“Cloudshine”=Scattered Ambient StarlightExtinction and scattering! (Cloudshine)
Almost Tasting a Very Simple Recipe
Foster & Goodman
2006
Extinction and scattering! (Cloudshine)
Theorists doing the Tasting!Extinction and scattering! (Cloudshine)
Padoan et al. 2006
Tastes “right”, with 20% scatter, at 1<AV<10, for NIR.
Recovered map
Simulation
“Cloudshine” Scattering
ModelH-band flux only
Meaningful structure in position-position-velocity space (3D)
Modeling l.o.s. temperature fluctuations
Is the gas density distribution lognormal or not? (2D)
And scattering! (Cloudshine models)
Spectral-Line Data: The 2D (bland)& the “3D” (Spicy) Views
Where am I?
What stars form from what gas, when?
position-postion-velocity is NOT the same asposition-position-position-velocity-velocity-velocitycf. Ostriker, Stone & Gammie 2001
mm peak (Enoch et al. 2006)
sub-mm peak (Hatchellet al. 2005, Kirk et al. 2006)
13CO (Ridge et al. 2006)
mid-IR IRAC composite from c2d data (Foster, Laakso, Ridge, et al. in prep.)
Optical image (Barnard 1927)
3D rendering courtesy AstroMed team at IIC (Borkin, Halle, Kauffmann, Alan, Goodman)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
0.40
0.35
0.30
0.25
0.20
Equivalent A
V [mag]
1-σ uncertainty (NICER)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
shellHD 278942 20
19
18
17
16
60 to 100µ C
olor Temperature [K
]
Dust Color Temperature (IRAS)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
2.0
1.5
1.0
0.5
0.0
τ ( 13CO
)
Opacity (13CO)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
~5 pc 12
10
8
6
4
2
Equivalent A
V [mag]
Extinction (NICER/2MASS)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
12
10
8
6
4
2
Equivalent A
V [mag]
Dust Emission (IRAS)
33
32
31
30
Dec
linat
ion
[J20
00, D
egre
es]
58 56 54 52 50
Right Ascension [J2000, Degrees]
12
10
8
6
4
2
Equivalent A
V [mag]
Line Emission (13CO)
Is the gas density distribution lognormal or not? (2D)
The (secret) uncertainties inherent in
column density mapping.
Goodman et al. 2007
Extin
ctio
nD
ust
Emis
sion
13C
O E
mis
sion
400
300
200
100
0
Num
ber
log (Equivalent AV [mag])
13COlognormal fit
to 13CO W(13CO )
400
300
200
100
0
Num
ber
2MASS/NICER
lognormal fit to2MASS column density
(all panels)
400
300
200
100
0
Num
ber
1.00.50.0
IRAS
lognormal fit toIRAS column density
Implied Column Density Distributions and lognormal Fits(Perseus COMPLETE data)
★Extinction & thermal emission are log-normal-ish & more so when non-13CO detected points are included.
★13CO is not a very faithful tracer of column density .
Is the gas density distribution lognormal or not? (2D)
Dendrograms (Hierarchical) vs. CLUMPFIND (Non-hierarchical)Meaningful structure in position-position-velocity space (3D)
Dendrogram (Rosolowsky et al. 2007;
cf. Houlahan & Scalo 1992)
CLUMPFIND(Williams et al. 1994)
Sky “x” (Right Ascension)
ObservedReality
“Observed”Simulations
(Dendro)Surfaces “CLUMPFIND”
work of Rosolowsky, Pineda, Kauffmann, Borkin,Padoan, Halle & Goodman; figure from Goodman & Rosolowsky NSF “Star Formation Taste Tests” Proposal, Fall 2006
Sky “y” (Declination
)
Velocity
Meaningful structure in position-position-velocity space (3D)
step 0.3 step 0.5threshold 0.3
threshold 0.5
threshold 0.7
Is CLUMPFIND OK as a Statistic? (Like Cayenne Pepper?)
CLUMPFIND output for L1448 with 1.2K step & threshold!(Lower values give too many clumps to show!!)
Rosolowsky, Borkin, Goodman, Kauffmann & Pineda 2007, in prep.
Dendrograms: A Physical Hierarchy?
The “self-gravitating” parts
of L1448:
And, coming soon, to a cookbook (Journal)
near you... (3D PDF)
Which “stars” “form” from what gas, when?Fi
gure
Cre
dit:
Jona
than
Fos
ter
L1448
'Cause if my eyes don't deceive me,There's something going wrong around here
--Joe Jackson
Theorists using
Observers Ingredients
e.g. Schmeja & Klessen 2006
What stars form from what gas, when?
Are you hungry yet?
Who can make this?