Tracing protostellar environments with H2O and CO:
from low to high mass with Herschel
Joseph Mo<ram
I. San José-‐García, A. Karska, L.E. Kristensen, E.F. van Dishoeck and the WISH and WILL survey teams
1
MoPvaPng quesPons
• What physical component(s) do H2O and CO trace and what are their properPes?
• Can we use LM sources as scaled templates for HM sources?
• Does this extend to cluster or extragalacPc scales?
2 2 2 R. Visser
The Data
3
WISH: Water In Star-‐forming regions with Herschel
• 425 hrs of Herschel Pme (van Dishoeck et al., 2011, PASP)
• HIFI spectroscopy & PACS spectral maps of H2O, CO and related molecules
• ~ 80 sources: – From 1 L¤ -‐ 105L¤
– Prestellar cores to disks
4
WILL: William Herschel Line Legacy
• OT2 follow-‐up to WISH-‐LM of a staPsPcally selected sample (Mo<ram et al., in prep.)
• ~50 sources selected from Herschel and Spitzer GB surveys using the criteria:
– α > 0.3 – Tbol < 300 K – Lbol > 0.4 L¤ for Class 0 – Lbol > 1 L¤ for Class I – δ < 35° for ALMA follow-‐up
– Vefng with HCO+ where available
5 101 102 103
Tbol (K)
100
101
102
Lbol(L
⊙)
• Class 0 • Class I
• WISH × WILL
Excellent data quality
6
Karska et al., 2014
Excellent data quality
7
Mo5ram et al., in prep.
0.0
0.1
0.2 AQU 01× 1.3
AQU 02× 1.5
AQU 03× 5
AQU 04× 5
AQU 05× 4
0.0
0.1
0.2 AQU 06× 5
CHA 01× 5
CHA 02× 5
CRA 01× 4
OPH 01× 5
0.0
0.1
0.2 OPH 02× 2
PER 01× 2
PER 02× 1.5
PER 03× 0.6
PER 04× 5
0.0
0.1
0.2 PER 05× 3
PER 06× 1.5
PER 07× 2
PER 08 PER 09
0.0
0.1
0.2 PER 10× 2
PER 11× 0.3
PER 12× 0.8
PER 13× 2
PER 14× 5
−75 0 75
v(km s−1)
0.0
0.1
0.2
TMB(K
) PER 15× 3
−75 0 75
PER 16× 5
−75 0 75
PER 17× 5
−75 0 75
PER 18× 5
H2O 110 − 101557 GHz
H2O 110-‐101 557GHz
Low-‐mass protostars – simplest template of star
formaPon
8
MulP-‐component H2O line profiles
• H2O line profiles are complex -‐> trace mulPple kinemaPc components
• Dominated by broad component associated with ounlows and shocks
• The FWHM is larger than for low-‐J CO
9
0.0
0.2
0.4110−101
−70 0 70
v(km s−1)
0.0
0.2
0.4
TMB(K
) × 1.5312−221
× 2111−000
SS
× 3202−111
C
−70 0 70
× 3211−202
S
Kristensen et al., 2012, Mo5ram et al., 2013, 2014
BHR71 Class 0 14.8 L¤ Menv=3.1M¤
200 pc
MulP-‐component CO rotaPonal diagram
• CO rotaPonal diagrams from PACS show a warm (~300K) mid-‐J and somePmes also a hot (~750K) high-‐J component
• Temperatures similar for all sources
10
Karska et al., 2013, 2014b see also Green et al., 2013, Manoj et al., 2013
• Water and low-‐J CO trace different spaPal regions within the ounlow
• Water limited to central PACS spaxel in bulk of sources; few show extended emission.
SpaPal extent of water
11
Tafalla et al., 2013, Karska et al., in prep.
• Cavity shock – C-‐shock along ounlow cavity
wall (central H2O, warm mid-‐J CO)
– Dominates H2O integrated intensity
– Tgas~300 K
• Spot Shocks – J-‐shock internal to jet and at
point of first impact on cavity wall (offset/EHV H2O, hot high-‐J CO)
– Tgas~750 K
Origin of the emission
12 12
R. Visser
Mo5ram et al., 2014 building on Kristensen et al., 2012, 2013, Karska et al., 2013
Water excitaPon • Lines are opPcally thick but effecPvely thin • Emifng region sizes are small, of order 10-‐200AU • n= 105-‐108 cm-‐3, N=1016-‐1018 cm-‐2
• RadiaPve pumping ruled out
13
12 13 14 15 16 17 18 19 20log(N [H2O]) (cm
−2)
2345678910
log(n[H
2])(cm
−3 )
50AU
100AU
250
AU
500AU
1000AU
50 100 150 200 250 300Eu/kb(K)
10−5
10−4
10−3
10−2
10−1
100
101
I/I 988
GHz
50 100 150 200 250 300Eu/kb(K)
10−5
10−4
10−3
10−2
10−1100101102103
τ
Two soluPons: marginally and fully sub-‐thermal
1016 1017 1018 1019 1020
N [H2O] (cm−2)
103
104
105
106
107
108
n[H
2](cm
−3 )
Cavity shock
Spot shock
Mo5ram et al., 2014
• Good agreement with shock models for single species PACS raPos
• Poor fit for inter-‐species raPos e.g. CO/H2O -‐> too much H2O in models
• H2O/CO 16-‐15 from HIFI finds X(H2O)~10-‐5
Need for UV irradiaPon
14 Karska et al., 2014, Kirstensen et al., in prep.
From low to high mass – does it all just scale?
15
• Profiles similar between Class 0, intermediate and high-‐mass WISH sources
• SPll dominated by cavity-‐shock component
Line profiles over probed L range
16 San Jose-‐Garcia et al., in prep.
Higher Lbol
• Integrated intensity of water and CO scale linearly with Lbol
• Warm PACS CO component also seen in HM sources with similar Trot
Line luminosiPes scales with L
17 San Jose-‐Garcia et al., 2013 & in prep., Karska et al., 2014 & in prep.
10-1 100 101 102 103 104 105 106
Lbol [LO • ]
10-4
10-3
10-2
10-1
100
101
102
L CO [K
km
s-1 p
c2 ]H2O CO
H2O vs. 12CO
• The FWZI (and FWHM) of H2O lines doesn’t vary with Lbol
• They increase with Lbol for CO • Therefore the relaPonship between H2O and 12CO changes with Lbol
• Also seen when comparing line raPos of H2O with 12CO J=10-‐9 and 16-‐15 as a funcPon of velocity
18 San Jose-‐Garcia et al., in prep.
H2O vs. CO – explanaPon(s)
• Two scenarios to explain this: – 1: Higher UV in HM sources destroys water closer to the cavity but releases more water further in
– 2: Higher turbulence caused by the ounlow leads to mixing of material deeper into the cavity wall
• Taken together, ounlows seem to scale with Lbol
19
Water emissionTurbulence
LM
Disk
Jet
Cavity wallOutflow cavity 3-
2
6-5
10-9
16-15
San Jose-‐Garcia et al., in prep.
From the Milky Way to other Galaxies
20
• H2O and CO intensity seem to conPnue to scale with Lbol to extragalacPc scales
• CombinaPon of LM and HM sources in beam?
Can we keep on scaling up?
21 San Jose-‐Garcia et al., 2013 & in prep., Kristensen & Bergin, in prep.
100 102 104 106 108 1010 1012 1014
Lbol [LO • ]
10-5
100
105
1010
L CO [K
km
s-1 p
c2 ]
ExtragalacUc data from Yang et al., 2013, van der Werf et al., 2010, Spinoglio et al., 2012, Kamenetzky et al. 2012; Meijerink et al. 2013
H2O
• ExtragalacPc water line-‐raPos more similar to LM than HM over common energy range
• Hot CO from protostars in extragalacPc sources too?
What about excitaPon?
22 San Jose-‐Garcia et al. in prep.
Conclusions
23
Conclusions
• Water and warm (~300K, mid-‐J) CO are dominated by the compact cavity shock in both LM and HM sources
• Integrated intensity scales linearly with Lbol • ExcitaPon of warm CO consistent between LM and HM, H2O under invesPgaPon
• A{er accounPng for details, LM can be scaled up to HM (at least for the ounlow physics we are probing)
• LM and HM sources provide templates which may be able explain observed extragalacPc emission
24
More details and papers can be found at h<p://www.strw.leidenuniv.nl/WISH/
Thank you for your a<enPon.
Any quesPons? 25