Characterizing the Molecular Interstellar Medium in ... - Caltech
Characterizing the Molecular Interstellar Medium in ... - Caltech
Characterizing the Molecular Interstellar Medium in ... - Caltech
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
<strong>Characteriz<strong>in</strong>g</strong> <strong>the</strong> <strong>Molecular</strong><br />
<strong>Interstellar</strong> <strong>Medium</strong> <strong>in</strong> Galaxies us<strong>in</strong>g<br />
Archival Herschel Data<br />
Naseem Rangwala<br />
University of Colorado - Boulder<br />
NASA Ames Research Center<br />
Ma<strong>in</strong> Collaborators: Jason Glenn (CU)<br />
Julia Kamenetzky (CU)<br />
Phil Maloney (CU)<br />
This work is supported by a NASA ADAP grant
•<br />
•<br />
•<br />
•<br />
•<br />
•<br />
Outl<strong>in</strong>e<br />
Description of <strong>the</strong> program<br />
Goals<br />
Sample<br />
Methodology/Pipel<strong>in</strong>e<br />
Prelim<strong>in</strong>ary Results (spectra, model<strong>in</strong>g results,<br />
CO SLEDs etc.)<br />
Conclusions
•<br />
Description of our Archival Program<br />
Archival (publicly available) survey us<strong>in</strong>g photometric and<br />
spectroscopic data of all <strong>the</strong> galaxies (~300) observed with Herschel-<br />
SPIRE’s Fourier Transform Spectrometer (FTS).
Flux [Jy]<br />
•<br />
•<br />
50<br />
40<br />
30<br />
20<br />
10<br />
CO (43)<br />
CI (10)<br />
CH<br />
CH<br />
CO (54)<br />
CO (65)<br />
HCN 87<br />
HCO+ 87<br />
H2O +<br />
H2O +<br />
H2O 211202<br />
HCN 98<br />
HCO+ 98<br />
CO (76)<br />
CI (21)<br />
CH +<br />
HCN 109<br />
HCO+ 109<br />
NH2 OH +<br />
CO (87)<br />
NH2 OH +<br />
HCN 1110<br />
NH<br />
HCO+ 1110<br />
H2O 202111<br />
NH<br />
OH +<br />
CO (98)<br />
HCN 1211<br />
HCO+ 1211<br />
H2O 312303<br />
H2O 111000<br />
H2O +<br />
H2O +<br />
HCN 1312<br />
CO (109)<br />
H2O 312221<br />
HCO+ 1312<br />
H2O 321312<br />
H2O 422413<br />
NH3 H2O 220211<br />
HF<br />
HCN 1413<br />
HCO+ 1413<br />
CO (1110)<br />
HCN 1514<br />
HCO+ 1514<br />
CO (1211)<br />
H2O 523514<br />
HCN 1615<br />
HCO+ 1615<br />
NH2 NH2 NII<br />
CH<br />
CH<br />
CO (1312)<br />
HCN 1716<br />
HCO+ 1716<br />
0<br />
670 µm<br />
SPIRE-FTS spectrum of Arp 220<br />
300 µm<br />
600 800 1000<br />
rest [GHz]<br />
1200 1400<br />
Cont<strong>in</strong>uous spectral coverage<br />
spectral resolution = 1.44 GHz<br />
•<br />
•<br />
Rangwala et al. (2011)<br />
several molecular and atomic species<br />
190 µm<br />
CO rotational transitions from J = 4-3 to 13-12
•<br />
•<br />
Description of our Archival Program<br />
Archival (publicly available) survey us<strong>in</strong>g photometric and<br />
spectroscopic data of all <strong>the</strong> galaxies (~300) observed with Herschel-<br />
SPIRE’s Fourier Transform Spectrometer (FTS).<br />
This sample spans a wide range <strong>in</strong> <strong>the</strong> far-<strong>in</strong>frared lum<strong>in</strong>osity (L FIR)<br />
and galaxy types.
N<br />
SPIRE- FTS Archival Sample<br />
80<br />
60<br />
40<br />
20<br />
Full Sample<br />
with CO J = 76<br />
LIRG<br />
0<br />
8 9 10 11<br />
Log LFIR (Lsun) 12 13 14<br />
•Ntot<br />
~ 300 (AGN = 78, Ellipticals = 32)<br />
•168/227<br />
galaxies have CO J = 7-6 detection<br />
•redshift<br />
up to z = 0.2<br />
ULIRG<br />
HLIRG
•<br />
•<br />
•<br />
•<br />
Description of our Archival Program<br />
Archival (publicly available) survey us<strong>in</strong>g photometric and<br />
spectroscopic data of all <strong>the</strong> galaxies (~300) observed with Herschel-<br />
SPIRE’s Fourier Transform Spectrometer (FTS).<br />
This sample span a wide range <strong>in</strong> <strong>the</strong> far-<strong>in</strong>frared lum<strong>in</strong>osity (L FIR)<br />
and galaxy types.<br />
CO spectral l<strong>in</strong>e energy distributions (SLEDs) from <strong>the</strong>se spectra<br />
will be uniformly re-reduced, re-calibrated and modeled to estimate<br />
<strong>the</strong> physical conditions and reservoir of <strong>the</strong> molecular gas<br />
Application of consistent methodology for data reduction and<br />
model<strong>in</strong>g across <strong>the</strong> entire sample will enable accurate comparison<br />
between different galaxies
•<br />
•<br />
•<br />
•<br />
•<br />
•<br />
Goals of <strong>the</strong> Program<br />
Determ<strong>in</strong>e <strong>the</strong> molecular gas (and dust) properties as a function of<br />
L FIR and galaxy type (starburst, AGN, disks and ellipticals)<br />
Address <strong>the</strong> orig<strong>in</strong> of <strong>the</strong> excitation of warm molecular gas of<br />
vary<strong>in</strong>g L FIR: Star-Formation vs AGN?<br />
Compare <strong>the</strong> gas excitation and dust properties of <strong>the</strong> local IR<br />
lum<strong>in</strong>ous galaxies to high-z submm galaxies.<br />
Explore *Total* L CO - L FIR Relation<br />
Direct measurement of [CO/H 2] abundance for warm molecular<br />
gas.<br />
Substantial public database of legacy value for nearby galaxies
Normalized Lum<strong>in</strong>osity<br />
10 7<br />
10 6<br />
CO SLED: diagnostic for<br />
Star Formation Vs AGN?<br />
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14<br />
J upper<br />
Mrk 231 (AGN)<br />
Arp 220 (Starburst)<br />
M82 (Starburst)<br />
Mid- to High-J CO traces warm gas and dom<strong>in</strong>ates <strong>the</strong> CO lum<strong>in</strong>osity and cool<strong>in</strong>g
•<br />
•<br />
•<br />
•<br />
•<br />
•<br />
Goals of <strong>the</strong> Program<br />
Determ<strong>in</strong>e <strong>the</strong> molecular gas (and dust) properties as a function of<br />
L FIR and galaxy type (starburst, AGN, disks and ellipticals)<br />
Address <strong>the</strong> orig<strong>in</strong> of <strong>the</strong> excitation of warm molecular gas of<br />
vary<strong>in</strong>g L FIR: Star-Formation Vs AGN?<br />
Compare <strong>the</strong> gas excitation and dust properties of <strong>the</strong> local IR<br />
lum<strong>in</strong>ous galaxies to high-z submm galaxies.<br />
Explore *Total* L CO - L FIR Relation<br />
Direct measurement of [CO/H 2] abundance for warm molecular<br />
gas.<br />
Substantial public database of legacy value for nearby galaxies
Herschel!<br />
Archive!<br />
Literature!<br />
data!!<br />
!<br />
Pipel<strong>in</strong>e Schematic<br />
SPIRE!<br />
Photometry!<br />
and!<br />
Spectrum!<br />
Photometry!data:!!<br />
10!–!850!µm!<br />
Dust!SED!Model<strong>in</strong>g!<br />
and!Likelihood!<br />
analysis!<br />
Probability!<br />
Distributions:!!<br />
Tdust,''τ,'&'β'<br />
Dust:!<br />
Temperature,!optical!<br />
depth,!FIR!<br />
lum<strong>in</strong>osity!and!Mass!<br />
Spectrum:!<br />
Additional!<br />
Reprocess<strong>in</strong>g!for!<br />
extended!sources!<br />
Literature!<br />
lowJJ!CO!<br />
ext<strong>in</strong>ction correction<br />
L<strong>in</strong>e!Fitter!!<br />
(Integrated!<br />
Fluxes!<strong>in</strong>!Jy!<br />
km/s!)!<br />
CO!SLEDs!!<br />
J!=!1J0!to!!<br />
J!=!13J12!<br />
Model<strong>in</strong>g!CO!and!<br />
Likelihood!<br />
analysis!<br />
Probability!<br />
Distributions:!T,#<br />
n,#P,!NCO,$&$Mgas$<br />
<strong>Molecular</strong>!Gas:!<br />
Temperature!<br />
Density!<br />
Mass!<br />
Lum<strong>in</strong>osity!<br />
Pipel<strong>in</strong>e developed by<br />
J. Kamenetzky<br />
L CO /L FIR<br />
CO#Excitation:<br />
PDR,#cosmic<br />
rays,#XDR,<br />
Mechanical<br />
energy
DATA
Flux (Jy)<br />
600<br />
400<br />
200<br />
0<br />
CO (43)<br />
CI (10)<br />
CO (54)<br />
CO (65)<br />
M82 (Starburst)<br />
CO (76)<br />
CI (21)<br />
CO (87)<br />
CO (98)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0008]<br />
This survey will focus only on CO, CI and NII l<strong>in</strong>es. CO and CI for<br />
characteriz<strong>in</strong>g molecular gas and NII as a star formation tracer<br />
CO (109)<br />
CO (1110)<br />
CO (1211)<br />
NII 205<br />
CO (1312)<br />
CO (1413)
Flux (Jy)<br />
12<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
CO (43)<br />
CI (10)<br />
CO (54)<br />
NGC 6240 (LIRG/AGN)<br />
CO (65)<br />
CO (76)<br />
CI (21)<br />
CO (87)<br />
CO (98)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0243]<br />
CO (109)<br />
CO (1110)<br />
CO (1211)<br />
NII 205<br />
CO (1312)<br />
CO (1413)
Flux (Jy)<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
CO (43)<br />
CI (10)<br />
CO (54)<br />
MRK 231 (ULIRG/AGN)<br />
CO (65)<br />
CO (76)<br />
CI (21)<br />
CO (87)<br />
CO (98)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0418]<br />
CO (109)<br />
CO (1110)<br />
CO (1211)<br />
NII 205<br />
CO (1312)<br />
CO (1413)
UGC 05101<br />
(LIRG/AGN)<br />
Flux (Jy)<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
CO (43)<br />
CI (10)<br />
CO (54)<br />
CO (65)<br />
CO (76)<br />
CI (21)<br />
CO (87)<br />
CO (98)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0393]<br />
CO (109)<br />
CO (1110)<br />
CO (1211)<br />
NII 205<br />
CO (1312)<br />
CO (1413)
Model<strong>in</strong>g and<br />
Prelim<strong>in</strong>ary Results
ladder<br />
FTS<br />
Ground<br />
JCMT<br />
Z−spec<br />
10 12 14<br />
–47–<br />
I (K km s −1 )<br />
10 5<br />
10 4<br />
10 3<br />
10 2<br />
Non-LTE Radiative Transfer Model<strong>in</strong>g Arp<br />
this work<br />
ground−based<br />
T = 1350 K, V = 350 km s −1<br />
T = 50 K, V = 500 km s −1<br />
0 5 10 15<br />
J upper<br />
cted lum<strong>in</strong>osity distribution of <strong>the</strong> CO ladder from J = 1-0<br />
of J = 10-9 l<strong>in</strong>e, which is blended withawaterl<strong>in</strong>e. The<br />
urements and <strong>the</strong> blue triangles are average l<strong>in</strong>e fluxes from<br />
lso shown for comparison are new measurements by Z-spec<br />
del fit to <strong>the</strong> observed CO l<strong>in</strong>e temperatures. There are two<br />
m component (dotted l<strong>in</strong>e) traced by <strong>the</strong> mid-J to high-J<br />
hed l<strong>in</strong>e) traced by <strong>the</strong> low-J l<strong>in</strong>es.<br />
<br />
%$ # <br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
•<br />
•<br />
•<br />
<br />
% # <br />
10.0<br />
Mid- to high-J CO are trac<strong>in</strong>g warm<br />
1.0<br />
<br />
<br />
<br />
<br />
%# <br />
<br />
<br />
del<strong>in</strong>g of CO: Likelihood distributions for gas k<strong>in</strong>etic temand<br />
pressure. The blue l<strong>in</strong>e represents <strong>the</strong> cold component<br />
g low-J transitions and <strong>the</strong> red l<strong>in</strong>e represents <strong>the</strong> warm<br />
from model<strong>in</strong>g <strong>the</strong> mid-J to high-J transitions.<br />
•<br />
•<br />
gas<br />
low-J CO are trac<strong>in</strong>g cold molecular<br />
gas.<br />
ARP220 − CO ladder<br />
0 2 4 6 8 10 12 14<br />
J upper<br />
T k<strong>in</strong>(warm CO) = T k<strong>in</strong>(warm H 2)<br />
M gas(warm) ~ 10% M gas(cold)<br />
<br />
<br />
<br />
L <br />
CO (cold) ~ 8% Total L CO<br />
<br />
! $<br />
<br />
L [x 10 6 L Sun]<br />
0.1<br />
FTS<br />
Ground<br />
JCMT<br />
Z−spec<br />
–47–<br />
220<br />
L [x 10 6 L Sun]<br />
10.0<br />
1.0<br />
0.1<br />
ARP220 − CO ladder<br />
FTS<br />
Ground<br />
JCMT<br />
Z−spec<br />
0 2 4 6 8 10 12 14<br />
J upper<br />
–47–<br />
I (K km s −1 )<br />
10 5<br />
10 4<br />
10 3<br />
10 2<br />
this work<br />
ground−based<br />
T = 1350 K, V = 350 km s −1<br />
T = 50 K, V = 500 km s −1<br />
0 5 10 15<br />
J upper<br />
Fig. 5.— Left: Ext<strong>in</strong>ction corrected lum<strong>in</strong>osity distribution of <strong>the</strong> CO ladder from J = 1-0<br />
to J = 13-12 with <strong>the</strong> exception of J = 10-9 l<strong>in</strong>e, which is blended withawaterl<strong>in</strong>e. The<br />
black solid circles are FTS measurements and <strong>the</strong> blue triangles are average l<strong>in</strong>e fluxes from<br />
ground-based measurements. Also shown for comparison are new measurements by Z-spec<br />
and JCMT. Right: Non-LTE model fit to <strong>the</strong> observed CO l<strong>in</strong>e temperatures. There are two<br />
temperature components: a warm component (dotted l<strong>in</strong>e) traced by <strong>the</strong> mid-J to high-J<br />
l<strong>in</strong>es and a cold component (dashed l<strong>in</strong>e) traced by <strong>the</strong> low-J l<strong>in</strong>es.<br />
%&#" ' !" %%<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
! $<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
%$# <br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
%# <br />
<br />
<br />
<br />
<br />
<br />
<br />
%# <br />
<br />
<br />
Fig. 6.— Radiative transfer model<strong>in</strong>g of CO: Likelihood distributions for gas k<strong>in</strong>etic temperature,<br />
density, column density and pressure. The blue l<strong>in</strong>e represents <strong>the</strong> cold component<br />
at 50 K obta<strong>in</strong>ed from model<strong>in</strong>g low-J transitions and <strong>the</strong> red l<strong>in</strong>e represents <strong>the</strong> warm<br />
component at 1350 K obta<strong>in</strong>ed from model<strong>in</strong>g <strong>the</strong> mid-J to high-J transitions.<br />
I (K km s −1 )<br />
10 5<br />
10 4<br />
10 3<br />
10 2<br />
this work<br />
ground−based<br />
T = 1350 K, V = 350 km s −1<br />
T = 50 K, V = 500 km s −1<br />
0 5 10 15<br />
J upper<br />
Fig. 5.— Left: Ext<strong>in</strong>ction corrected lum<strong>in</strong>osity distribution of <strong>the</strong> CO ladder from J = 1-0<br />
to J = 13-12 with <strong>the</strong> exception of J = 10-9 l<strong>in</strong>e, which is blended withawaterl<strong>in</strong>e. The<br />
black solid circles are FTS measurements and <strong>the</strong> blue triangles are average l<strong>in</strong>e fluxes from<br />
ground-based measurements. Also shown for comparison are new measurements by Z-spec<br />
and JCMT. Right: Non-LTE model fit to <strong>the</strong> observed CO l<strong>in</strong>e temperatures. There are two<br />
temperature components: a warm component (dotted l<strong>in</strong>e) traced by <strong>the</strong> mid-J to high-J<br />
l<strong>in</strong>es and a cold component (dashed l<strong>in</strong>e) traced by <strong>the</strong> low-J l<strong>in</strong>es.<br />
%&#" ' !" %%<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
%$# <br />
•Radiative(Transfer(code(ma<strong>in</strong>ta<strong>in</strong>ed(by!Phil!Maloney<br />
cool<strong>in</strong>g rate = 22 L /M <br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
<br />
%# <br />
<br />
<br />
<br />
<br />
<br />
<br />
%# <br />
<br />
<br />
Fig. 6.— Radiative transfer model<strong>in</strong>g of CO: Likelihood distributions for gas k<strong>in</strong>etic temperature,<br />
density, column density and pressure. The blue l<strong>in</strong>e represents <strong>the</strong> cold component<br />
at 50 K obta<strong>in</strong>ed from model<strong>in</strong>g low-J transitions and <strong>the</strong> red l<strong>in</strong>e represents <strong>the</strong> warm<br />
component at 1350 K obta<strong>in</strong>ed from model<strong>in</strong>g <strong>the</strong> mid-J to high-J transitions.<br />
Rangwala et al. (2011)
Arp 220: Excitation source of warm<br />
molecular gas<br />
•<br />
•<br />
•<br />
•<br />
Observed ratio: Total L CO/L FIR ~ 10 -4<br />
[T, n(H 2)] cold = [50 K, 1000 cm -3 ] and [T, n(H 2)] warm = [1350<br />
K, 1000 cm -3 ]<br />
This ratio along with tight constra<strong>in</strong>ts on T k<strong>in</strong> and n(H 2)<br />
rules out PDRs, XDRs and Cosmic ray models<br />
Require a non-ioniz<strong>in</strong>g source of energy: A small fraction<br />
of mechanical energy from supernovae and stellar w<strong>in</strong>ds<br />
can satisfy a cool<strong>in</strong>g rate = 22 L /M <br />
Similar method has been successfully applied <strong>in</strong> cases NGC 1068 (Sp<strong>in</strong>oglio et<br />
al. 2012; Hailey-Dunsheath et al. 2012), M82 (Kamenetzky et al. 2012), NGC<br />
4038/39 (Schirm et al. <strong>in</strong> prep.) and Cloverleaf (Bradford et al. 2009)
3 4 5 6<br />
Log 10 n(H 2) (cm 3 )<br />
5 6 7 8<br />
Log 10 Pressure (K cm 3 )<br />
I CO (K km s 1 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
18 19 20 21 22 23<br />
Log10 NCO (cm 2 0.0<br />
)<br />
10000<br />
1000<br />
2<br />
Log10 N[H2] [cm ]<br />
23 24 25 26 27 28<br />
0 2 4 6 8 10 12 14<br />
J upper<br />
•<br />
•<br />
•<br />
Cold<br />
Warm<br />
Total<br />
Prelim<strong>in</strong>ary<br />
NGC6240: CO Model<strong>in</strong>g<br />
T k<strong>in</strong>(warm CO) ~ 1580 K<br />
T k<strong>in</strong>(cold CO) ~ 50<br />
0.2<br />
0.2<br />
cool<strong>in</strong>g rate ~ 35 L/M 0.0 0.0<br />
10 100 1000<br />
Log10 Tk<strong>in</strong> (K)<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
Cold<br />
Warm<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
2 3 4 5 6<br />
Log 10 n(H 2) (cm 3 )<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
Cold<br />
Warm<br />
10 100 1000<br />
Log 10 T k<strong>in</strong> (K)<br />
0.0<br />
1.0 0.8 0.6 0.4 0.2 0.0<br />
Log10 ff<br />
2<br />
Log10 N[H2] [cm ]<br />
23 24 25 26 27 28<br />
18 19 20 21 22 23<br />
Log10 NCO (cm 2 0.0<br />
)<br />
1.0 1.0 Cold<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
2 3 4 5 6<br />
Log 10 n(H 2) (cm 3 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
3 4 5 6 7 8<br />
Log 10 Pressure (K cm 3 )<br />
Marg<strong>in</strong>alized Likellihood<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
Cold<br />
Warm<br />
10 100 1000<br />
Log 10 T k<strong>in</strong> (K)<br />
0.0<br />
1.0 0.8 0.6 0.4 0.2 0.0<br />
Log10 ff<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
I CO (K km s 1 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
2<br />
Log10 N[H2] [cm ]<br />
23 24 25 26 27 28<br />
18 19 20 21 22 23<br />
Log10 NCO (cm 2 0.0<br />
)<br />
10000<br />
1000<br />
2 3 4 5 6<br />
Log10 n(H2) (cm 3 0.0<br />
)<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
3 4 5 6 7 8<br />
Log10 Pressure (K cm 3 0.0<br />
)<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
Cold<br />
Warm<br />
Total<br />
0 2<br />
0.6<br />
4 6 8<br />
Jupper 10 12 14<br />
I CO (K km s 1 )<br />
0.4<br />
0.2<br />
2<br />
Log10 N[H2] [cm ]<br />
23 24 25 26 27 28<br />
18 19 20 21 22 23<br />
Log10 NCO (cm 2 0.0<br />
)<br />
10000<br />
1000<br />
Cold<br />
Warm<br />
Total<br />
0 2 4 6 8 10 12 14<br />
J upper
3 4 5 6<br />
Log 10 n(H 2) (cm 3 )<br />
4 5 6 7 8<br />
Log 10 Pressure (K cm 3 )<br />
I CO (K km s 1 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
1000<br />
100<br />
10<br />
•<br />
•<br />
•<br />
2<br />
Log10 N[H2] [cm ]<br />
21.0 21.5 22.0 22.5 23.0 23.5<br />
17.5 18.0 18.5 19.0 19.5 20.0<br />
Log 10 N CO (cm 2 )<br />
Cold<br />
Warm<br />
Total<br />
0 2 4 6 8 10 12 14<br />
J upper<br />
Prelim<strong>in</strong>ary<br />
M82: CO Model<strong>in</strong>g Results<br />
T k<strong>in</strong> (warm CO) ~ 500 K = T k<strong>in</strong> (warm H 2 )<br />
T k<strong>in</strong>(cold CO) ~ 35 K<br />
cool<strong>in</strong>g rate ~ 1.4 L /M <br />
Marg<strong>in</strong>alized Likellihood<br />
ikellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
Cold<br />
Warm<br />
10 100 1000<br />
Log 10 T k<strong>in</strong> (K)<br />
Marg<strong>in</strong>alized Likellihood<br />
ikellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
2 3 4 5 6<br />
Log10 n(H2) (cm 3 0.0<br />
)<br />
1.0<br />
0.8<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
Cold<br />
Warm<br />
10 100 1000<br />
Log 10 T k<strong>in</strong> (K)<br />
0.0<br />
3.0 2.5 2.0 1.5 1.0 0.5 0.0<br />
Log10 ff<br />
s 1 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1000<br />
100<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
2<br />
Log10 N[H2] [cm ]<br />
21.0 21.5 22.0 22.5 23.0 23.5<br />
17.5 18.0 18.5 19.0 19.5 20.0<br />
Log 10 N CO (cm 2 )<br />
Cold<br />
Warm<br />
Total<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
2 3 4 5 6<br />
Log10 n(H2) (cm 3 0.0<br />
)<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
0.8<br />
0.8<br />
3 4 5 6 7 8<br />
Marg<strong>in</strong>alized Likellihood<br />
Log10 Pressure (K cm 3 )<br />
0.6<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.4<br />
0.2<br />
0.0<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
Cold<br />
Warm<br />
10 100 1000<br />
Log 10 T k<strong>in</strong> (K)<br />
0.0<br />
3.0 2.5 2.0 1.5 1.0 0.5 0.0<br />
Log10 ff<br />
Marg<strong>in</strong>alized Likellihood<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
0.6<br />
0.4<br />
0.2<br />
I CO (K km s 1 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1000<br />
100<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
0.0<br />
10<br />
2 3 4 5 6<br />
Log10 n(H2) (cm 3 0.0<br />
)<br />
1.0<br />
0.8<br />
0.6<br />
0.4<br />
0.2<br />
3 4 5 6 7 8<br />
Log10 Pressure (K cm 3 0.0<br />
)<br />
2<br />
Log10 N[H2] [cm ]<br />
21.0 21.5 22.0 22.5 23.0 23.5<br />
17.5 18.0 18.5 19.0 19.5 20.0<br />
Log 10 N CO (cm 2 )<br />
Marg<strong>in</strong>alized Likellihood<br />
1.0<br />
Cold<br />
Warm<br />
Total<br />
0.8<br />
0 2 4 6 8 10 12 14<br />
Jupper 0.6<br />
I CO (K km s 1 )<br />
1000<br />
100<br />
0.4<br />
0.2<br />
0.0<br />
10<br />
2<br />
Log10 N[H2] [cm ]<br />
21.0 21.5 22.0 22.5 23.0 23.5<br />
17.5 18.0 18.5 19.0 19.5 20.0<br />
Log 10 N CO (cm 2 )<br />
Cold<br />
Warm<br />
Total<br />
0 2 4 6 8 10 12 14<br />
J upper
CO SLEDs<br />
Normalized Lum<strong>in</strong>osity<br />
10 7<br />
10 6<br />
0 115 230 345 461 576 691 806 922 1037 1152 1267 1382 1497 1611<br />
Mrk 231 (AGN)<br />
UGC05101 (AGN)<br />
Arp 220 (ULIRG/starburst)<br />
M82 (starburst)<br />
NGC7552 (barred spiral)<br />
Milky Way<br />
Frequency (GHz)<br />
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14<br />
Jupper Us<strong>in</strong>g a large sample we will establish if <strong>the</strong> shape and brightness of<br />
<strong>the</strong> CO SLED <strong>in</strong> galaxies can be used as a diagnostic between dom<strong>in</strong>ant<br />
sources of energy affect<strong>in</strong>g <strong>the</strong>ir molecular ISM and star formation.
Dust Model<strong>in</strong>g<br />
Arp220<br />
NGC 6240<br />
S( )= Nbb(1 e<br />
0<br />
e hc<br />
kT 1<br />
)(c/ ) 3<br />
+ Npl ↵ e ( / c) 2
•<br />
•<br />
•<br />
•<br />
•<br />
Conclusions<br />
<strong>Molecular</strong> gas and dust survey of nearby galaxies us<strong>in</strong>g<br />
Herschel-SPIRE.<br />
Systematic re-process<strong>in</strong>g and model<strong>in</strong>g; Pipel<strong>in</strong>e is almost<br />
complete.<br />
Previous studies have shown that <strong>the</strong> high-J CO l<strong>in</strong>es are trac<strong>in</strong>g<br />
<strong>the</strong> warm molecular ISM and dom<strong>in</strong>ate <strong>the</strong> CO lum<strong>in</strong>osity and<br />
hence <strong>the</strong> cool<strong>in</strong>g.<br />
Observed L CO/L FIR comb<strong>in</strong>ed with physical properties of <strong>the</strong><br />
molecular gas will enable us to constra<strong>in</strong> <strong>the</strong> dom<strong>in</strong>ant power<br />
source (AGN or SF (PDR, XDR, shocks etc. )) <strong>in</strong> <strong>the</strong>se galaxies.<br />
Investigate global properties of <strong>the</strong> molecular gas and dust as<br />
function of L FIR and galaxy types.
Extra Slides
Arp 220: Excitation source of warm<br />
molecular gas<br />
•<br />
•<br />
•<br />
!44B444<br />
!4B444<br />
!B444<br />
4B!44<br />
4B4!4<br />
4B44!<br />
Observed ratio: Total L CO/L FIR ~ 10 -4<br />
This ratio along with tight constra<strong>in</strong>ts on T k<strong>in</strong> and n(H 2)<br />
rules out PDRs, XDRs and Cosmic rays<br />
Require a non-ioniz<strong>in</strong>g source of energy: Mechanical<br />
energy from supernovae and stellar w<strong>in</strong>ds to satisfy a<br />
cool<strong>in</strong>g rate = 20 L /M <br />
/E?@A/0)F7+/+ /12 /+F 3<br />
, -./01/23/4 & 5<br />
567/8 4<br />
(a)<br />
3456+,780<br />
&!!!!<br />
&<br />
%<br />
$<br />
#<br />
"<br />
!<br />
4<br />
%1"<br />
$12<br />
$1"<br />
"12<br />
"1"<br />
"12<br />
$1"<br />
9:+;& +;$ "=<br />
9:+;2 &=<br />
9:+;! 2=<br />
9:+;? !=<br />
9>+;% $=<br />
9:+;# ?=<br />
9:+;@ #=<br />
!"" #"" $""" $%"" $&"" $!""<br />
'()*+,-./0<br />
5(678-9:;*'-,:88-?67<br />
9:+;$$ $"=<br />
9:+;$% $$=<br />
A>>+; < B $ < B "=<br />
9:+;$< $%=<br />
UGC05101 - Two-Component Radiative Transfer Model<strong>in</strong>g<br />
Model Fit<br />
Temperature<br />
PDR Models<br />
567/9:0!" !!3;9:0% $3<br />
%B4<br />
ARP220<br />
$B4<br />
M82<br />
#B4<br />
-678<br />
9:+3<br />
;6
UGC 05101<br />
(LIRG/AGN)<br />
Flux [Jy]<br />
3456+,780<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
%1"<br />
$12<br />
$1"<br />
"12<br />
"1"<br />
"12<br />
$1"<br />
(a)<br />
9:+;&+;$"=<br />
(b)<br />
Flux (Jy)<br />
10<br />
8<br />
6<br />
4<br />
2<br />
0<br />
9:+;2&=<br />
CO (43)<br />
CI (10)<br />
CO (54)<br />
9:+;!2=<br />
CO (65)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0393]<br />
Herschel SPIRE-FTS Spectrum of UGC05101<br />
600 800 1000 1200 1400 1600<br />
rest [GHz]<br />
9:+;?!=<br />
9>+;%$=<br />
CO (76)<br />
CI (21)<br />
9:+;#?=<br />
CO (87)<br />
9:+;@#=<br />
CO (98)<br />
!"" #"" $""" $%"" $&"" $!""<br />
'()*+,-./0<br />
CO (109)<br />
9:+;$"@=<br />
CO (1110)<br />
9:+;$$$"=<br />
CO (1211)<br />
9:+;$%$$=<br />
NII 205<br />
CO (1312)<br />
A>>+; < B $ < B "=<br />
9:+;$
Normalized Lum<strong>in</strong>osity<br />
10 7<br />
10 6<br />
CO SLED: diagnostic for<br />
Star Formation Vs AGN?<br />
Mrk 231 (AGN)<br />
Arp 220 (Starburst)<br />
M82 (Starburst)<br />
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14<br />
Us<strong>in</strong>g a large sample Jupper we will establish if <strong>the</strong><br />
shape and brightness of <strong>the</strong> CO SLED <strong>in</strong><br />
galaxies can be used as a diagnostic between<br />
dom<strong>in</strong>ant sources of energy affect<strong>in</strong>g <strong>the</strong>ir<br />
molecular ISM and star formation.<br />
•<br />
•<br />
•<br />
Models have suggested that<br />
<strong>the</strong> shape of <strong>the</strong> CO SLEDs can<br />
be used to discrim<strong>in</strong>ate<br />
between SF and AGN [e.g. Spaans<br />
& Meijer<strong>in</strong>k (2008); Lagos et al. (2012)].<br />
Observations:<br />
Starburst - CO SLEDs turn<br />
over after J = 6-5 e.g.<br />
Arp220 [Rangwala et al. (2011)] and<br />
M82 [Panuzzo, Rangwala et al. (2010);<br />
Kamenetzky et al. 2012 (<strong>in</strong>cl. N. Rangwala)].<br />
Similar turn over observed <strong>in</strong><br />
high-z submm galaxies [Bothwell<br />
et al. (2013)].<br />
AGN: CO SLED rema<strong>in</strong>s flat<br />
after J = 6-5 [e.g.,Van der werf et al.<br />
(2010)]
L CO (65) (L sun)<br />
10 10<br />
10 9<br />
10 8<br />
10 7<br />
10 6<br />
10 5<br />
10 4<br />
cena<br />
10 10<br />
LCO - LFIR relation<br />
NGC1365NE<br />
NGC1365SW<br />
ngc1068<br />
ngc1266<br />
NGC253 m82 ngc4038overlap<br />
NGC1222 ngc4038<br />
m83<br />
10 11<br />
Mrk273<br />
arp220<br />
UGC05101<br />
Mrk231<br />
NGC6240<br />
IRAS090223615<br />
IRASF172070014<br />
Arp299A<br />
Arp299B Arp299C<br />
L FIR (L sun)<br />
10 12<br />
SMMJ2135<br />
10 13<br />
HLSW01<br />
Riechers<br />
GN20<br />
Cloverleaf<br />
APM08279<br />
10 14
log 10 L CO (65) (K km s 1 pc 2 )<br />
12<br />
11<br />
10<br />
9<br />
8<br />
7<br />
6<br />
cena<br />
L CO - L FIR relation<br />
SMMJ2135<br />
Mrk273<br />
arp220<br />
UGC05101<br />
NGC1365NE<br />
NGC1365SW<br />
ngc1068<br />
m82 ngc4038overlap<br />
m83 ngc4038<br />
Mrk231<br />
NGC6240<br />
IRAS090223615<br />
IRASF172070014<br />
Arp299A<br />
Arp299B Arp299C<br />
ngc1266<br />
NGC253<br />
NGC1222<br />
HLSW01 Cloverleaf<br />
Riechers<br />
GN20<br />
APM08279 MM18423<br />
5<br />
9 10 11 12<br />
log10 LFIR (Lsun) 13 14 15
3060 M. S. Bothwell et al.<br />
4.3 The L ′ CO –LFIR correlation<br />
A useful quantity to measure for our sample of SMGs is <strong>the</strong> efficiency<br />
with which <strong>the</strong>ir molecular gas is be<strong>in</strong>g converted <strong>in</strong>to stars.<br />
The SFE is sometimes def<strong>in</strong>ed as SFR/M(H2) – <strong>the</strong> <strong>in</strong>verse of <strong>the</strong> gas<br />
depletion time – but here we take <strong>the</strong> approach of parameteriz<strong>in</strong>g<br />
this as a ratio of observable quantities: LFIR and L ′ CO(1−0) .<br />
There has been debate, <strong>in</strong> recent years, as to <strong>the</strong> value of <strong>the</strong><br />
slope of <strong>the</strong> L ′ CO(1−0) − LFIR relation. Whereas, earlier we discussed<br />
<strong>the</strong> difference <strong>in</strong> slopes between <strong>the</strong> various transitions observed<br />
(<strong>in</strong> order to derive <strong>the</strong> median SLED), here we present <strong>the</strong> relation<br />
between <strong>the</strong> derived 12CO J = 1–0 lum<strong>in</strong>osity and LFIR – a relation<br />
which describes, <strong>in</strong> observable terms, <strong>the</strong> relationship between <strong>the</strong><br />
lum<strong>in</strong>osity due to star formation and <strong>the</strong> total gas content. (Section<br />
2.1.1 conta<strong>in</strong>s a discussion of <strong>the</strong> uncerta<strong>in</strong>ties <strong>in</strong>herent to deriv<strong>in</strong>g<br />
IR lum<strong>in</strong>osities for sources such as ours.)<br />
In <strong>the</strong>ir study of 12 SMGs, Greve et al. (2005) found a slope of<br />
<strong>the</strong> relation between LFIR and L ′ CO(1−0) of 0.62 ± 0.08 fit a comb<strong>in</strong>ed<br />
sample of lower redshift LIRGs, ULIRGs and SMGs – identical to<br />
<strong>the</strong> slope derived for <strong>the</strong> local LIRGs/ULIRGs alone. The small<br />
number of galaxies <strong>in</strong> Greve et al. (2005), however, prevented a full<br />
<strong>in</strong>vestigation of <strong>the</strong> SFE slopes with<strong>in</strong> <strong>the</strong> SMG population itself.<br />
Some recent authors, however, have found <strong>the</strong> slope to be closer to<br />
l<strong>in</strong>ear – Genzel et al. (2010) found that a slope of 0.87 ± 0.09 fits a<br />
comb<strong>in</strong>ed sample of SMGs across a wide range of redshifts.<br />
Fig. 6 shows <strong>the</strong> L ′ CO−LFIR relation for our sample of SMGs.<br />
Included <strong>in</strong> <strong>the</strong> plot are data po<strong>in</strong>ts for local (U)LIRGs, as measured<br />
by Sanders, Scoville & Soifer (1991) and Solomon et al.<br />
(1997). We also show three power-law fits to <strong>the</strong> local (U)LIRGs<br />
alone, <strong>the</strong> SMGs alone and <strong>the</strong> comb<strong>in</strong>ed sample. We f<strong>in</strong>d <strong>the</strong><br />
SMGs to lie slightly above <strong>the</strong> best-fitt<strong>in</strong>g (sub-l<strong>in</strong>ear) l<strong>in</strong>e for local<br />
(U)LIRGs, necessitat<strong>in</strong>g a steeper slope. The power-law fit to <strong>the</strong><br />
local (U)LIRGs alone has a slope of 0.79 ± 0.08, while <strong>the</strong> fit to<br />
<strong>the</strong> comb<strong>in</strong>ed sample of SMGs and (U)LIRGs has a slope of 0.83 ±<br />
0.09. It can also be seen that a fit to <strong>the</strong> SMG sample alone has<br />
an even steeper slope of 0.93 ± 0.14 – very close to l<strong>in</strong>ear – although,<br />
with<strong>in</strong> <strong>the</strong> uncerta<strong>in</strong>ty, this is consistent with <strong>the</strong> slope for<br />
<strong>the</strong> comb<strong>in</strong>ed samples. These results are <strong>in</strong> good agreement with<br />
Figure 6. The SFE, L ′ CO versus LFIR. Included <strong>in</strong> <strong>the</strong> plot are two samples<br />
of local ULIRGs, and <strong>the</strong> J = (1–0) SMG observations of Ivison et al.<br />
(2011). Best-fitt<strong>in</strong>g slopes to <strong>the</strong> SMGs alone, <strong>the</strong> local ULIRGs alone and<br />
all three comb<strong>in</strong>ed samples are overplotted. They have slopes of 0.93 ±<br />
0.14, 0.79 ± 0.08 and 0.83 ± 0.09, respectively.<br />
L CO - L FIR relation<br />
most previous f<strong>in</strong>d<strong>in</strong>gs; <strong>the</strong> near-l<strong>in</strong>ear slopes (which agree well<br />
with those found by Genzel et al. 2010) would imply a roughly<br />
constant gas depletion time-scale across <strong>the</strong> entire range of far-IR<br />
lum<strong>in</strong>osities shown here.<br />
However, we caution that our analysis requires extrapolat<strong>in</strong>g from<br />
high-Jup 12 CO transitions to 12 CO (1–0). Ivison et al. (2010) have<br />
undertaken a similar analysis based solely on directly observed<br />
12 CO (1–0) observations and homogeneously derived far-IR lumi-<br />
nosities and conclude that <strong>the</strong> L ′ CO(1−0) −LFIR relation has a slope<br />
substantially below unity. We <strong>the</strong>refore suggest that it is difficult to<br />
draw any strong conclusions from high-Jup observations about <strong>the</strong><br />
gas depletion time-scales of <strong>the</strong> different populations or <strong>the</strong> form of<br />
<strong>the</strong> Kennicutt–Schmidt relation <strong>in</strong> <strong>the</strong>se galaxies, as <strong>the</strong>se are too<br />
uncerta<strong>in</strong> without brightness temperature ratio measurements for<br />
<strong>in</strong>dividual sources.<br />
4.4 <strong>Molecular</strong> gas masses<br />
Part of <strong>the</strong> power of observations of 12CO emission from highredshift<br />
galaxies is that <strong>the</strong>y provide a tool to derive <strong>the</strong> mass of<br />
<strong>the</strong> reservoir of molecular gas <strong>in</strong> <strong>the</strong>se systems. This is of critical<br />
importance because this reservoir is <strong>the</strong> raw material from which<br />
<strong>the</strong> future stellar mass <strong>in</strong> <strong>the</strong>se systems is formed. Along with <strong>the</strong><br />
exist<strong>in</strong>g stellar population, it <strong>the</strong>refore gives some <strong>in</strong>dication of<br />
<strong>the</strong> potential stellar mass of <strong>the</strong> result<strong>in</strong>g galaxy at <strong>the</strong> end of <strong>the</strong><br />
starburst phase (subject, of course, to <strong>the</strong> unknown contribution<br />
from <strong>in</strong>-fall<strong>in</strong>g and out-flow<strong>in</strong>g material).<br />
Estimat<strong>in</strong>g <strong>the</strong> mass of H2 from <strong>the</strong> measured L ′ CO requires two<br />
steps. First, lum<strong>in</strong>osities orig<strong>in</strong>at<strong>in</strong>g from higher transitions (Jup ≥<br />
2) must be transformed to an equivalent 12CO J = 1–0 lum<strong>in</strong>osity,<br />
us<strong>in</strong>g a brightness ratio. We have derived <strong>the</strong> necessary brightness<br />
ratios us<strong>in</strong>g our composite SLED as discussed <strong>in</strong> Section 3 above.<br />
Once an L ′ CO(1−0) has been determ<strong>in</strong>ed, it must be converted <strong>in</strong>to<br />
an H2 mass by adopt<strong>in</strong>g a conversion factor α: M(H2) = αL ′ CO ,<br />
where α is <strong>in</strong> units of M⊙ (K km s−1 pc2 ) −1 (when discuss<strong>in</strong>g α<br />
hereafter, we omit <strong>the</strong>se units for <strong>the</strong> sake of brevity). This can <strong>the</strong>n<br />
be converted to a total gas mass, <strong>in</strong>clud<strong>in</strong>g He, Mgas = 1.36 M(H2).<br />
There is a large body of work, both observational and <strong>the</strong>oretical,<br />
dedicated to determ<strong>in</strong><strong>in</strong>g <strong>the</strong> value – and ascerta<strong>in</strong><strong>in</strong>g <strong>the</strong> metallicity<br />
or environmental dependence – of α (e.g. Young & Scoville<br />
1991; Solomon & Vanden Bout 2005; Liszt, Pety & Lucas 2010;<br />
Bolatto et al. 2011; Genzel et al. 2012; Narayanan et al. 2011; Papadopoulos<br />
et al. 2012). While secular discs such as <strong>the</strong> Milky Way<br />
have a relatively ‘high’ value of α ∼ 3–5, us<strong>in</strong>g this value for <strong>the</strong><br />
gas <strong>in</strong> nuclear discs/r<strong>in</strong>gs with<strong>in</strong> merg<strong>in</strong>g systems and starbursts at<br />
z ∼ 0 leads to <strong>the</strong> molecular gas mass sometimes exceed<strong>in</strong>g <strong>the</strong>ir<br />
dynamical masses. As such, a lower value – motivated by a radiative<br />
transfer model of <strong>the</strong> 12CO k<strong>in</strong>ematics – is typically used<br />
for <strong>the</strong> <strong>in</strong>tense nuclear starbursts <strong>in</strong> <strong>the</strong> most IR-lum<strong>in</strong>ous local<br />
systems: α ∼ 0.8, with a range of 0.3–1.3 (Downes & Solomon<br />
1998). However, some recent results have suggested that this value<br />
might, <strong>in</strong> fact, underestimate <strong>the</strong> true value <strong>in</strong> high-redshift SMGs.<br />
Bothwell et al. (2010) found that apply<strong>in</strong>g <strong>the</strong> canonical ULIRG<br />
value to two z ∼ 2 SMGs resulted <strong>in</strong> gas fractions of
Flux (Jy)<br />
•<br />
•<br />
50<br />
40<br />
30<br />
20<br />
10<br />
0<br />
670 µm<br />
CO (43)<br />
CI (10)<br />
SPIRE-FTS spectrum of Arp 220<br />
CO (54)<br />
CO (65)<br />
CO (76)<br />
CI (21)<br />
CO (87)<br />
CO (98)<br />
600 800 1000 1200 1400 1600<br />
rest (GHz) [used z = 0.0181]<br />
Cont<strong>in</strong>uous spectral coverage<br />
spectral resolution = 1.44 GHz<br />
•<br />
•<br />
CO (109)<br />
300 µm<br />
CO (1110)<br />
CO (1211)<br />
NII 205<br />
CO (1312)<br />
CO (1413)<br />
190 µm<br />
several molecular and atomic species<br />
CO rotational transitions from J = 4-3 to 13-12