Nearby H II regions and a far away one: the simplest models that fit them all

1
Nearby H II regions (and a far away one) the
simplest models that fit them all

• Mónica Rodríguez (INAOE, Mexico)?
• Jorge García-Rojas (UNAM, Mexico)?

2
What is the value of O/H (Z) in the solar
neighborhood? (and everywhere else?)?
? What can H II regions tell us? The best spectra
García-Rojas et al. (2004, 2005, 2006, 2007)
Esteban et al. (2004). Plus 30 Doradus (Peimbert
2003) in the LMC.
3

• ADF(O) (OREC)/(OCEL) 1.42.8
  
  
• Extremely difficult to reproduce with temperature
  fluctuations arising from well known processes.
  So
  
• some extra heating mechanism? (OREC
  Oreal)?
  
• metal-rich inclusions?
  (OCEL Oreal Stasinska et al.
  2007)?
  
• errors in the recombination coefficients?
  (OCEL Oreal)?
  
• Each explanation will produce a different Te
  structure
  
• Can one reproduce easily the temperatures and
  abundances implied by the collisionally excited
  lines using simple photoionization models?

4
The grid of models

• Constant density (n(H) 300, 1000, and 5000
  cm?3) plane-parallel photoionization models with
  Cloudy (Ferland et al. 1998)?
  
• ionized by a source with Teff 3500045000 K,
  log(g) 4 (WM-basic Pauldrach et al. 2001) and
  different numbers of H ionizing photons
  
• Orion type dust (silicates graphite, no small
  grains)?
  
• Orion abundances, scaled (along with the
  grains) 12log(O/H)in 8.39.1 (see Stasinska
  2005)?
  
• Models and observations analized in the same way
  and with the same atomic data ne, Te(N II)out
  ? (O/H)out, Te(O III)out ? (O/H)out,
  (O/H)out (O/H)out (O/H)out

plus some considerations
5

• effect of mixed Teff's
• effect of intervening material in the radiation
  field
  
• adding small grains
• open vs. closed geometry
• density laws
• changing the relative abundances of C, N, O,
  S...
  
• advection (see Henney et al. 2005)?
• other model atmospheres
  (see Simón-Díaz Stasinska 2008
  for a comparison)?

but these things lead to small changes in the
derived parameters, and we are not trying to
build realistic models of the individual objects,
just to capture their main characteristics
6
M17 M20 (ne 300 cm?3)? 12log(O/H)in 8.5 (s
mall symbols),8.6 (large ones)?
12log(O/H)out 8.448.58 12log(O/H)observed
8.48, 8.53
recombination contribution to N II ?5754
M17
M17
M20
M20
Rubin (1986) Liu et al. (2000)?
7
NGC 3576, M8 M16 (ne 1000 cm?3)?
12log(O/H)in 8.5 (small symbols),8.6 (large
ones)? 12log(O/H)out 8.448.58 12log(O/H)ob
served 8.44, 8.52, 8.53
NGC 3576
NGC 3576
M8
M8
M16
M16
8
NGC 3603 M2 (ne 5000 cm?3)?
12log(O/H)in 8.5 (small symbols),8.6 (large
ones)? 12log(O/H)out 8.438.56 12log(O/H)ob
served 8.48, 8.52
NGC 3603
M42
NGC 3603
M42
9
30 Doradus S311 (ne 300 cm?3)?
12log(O/H)in 8.3 (small symbols),8.4 (large
ones)? 12log(O/H)out 8.268.39 12log(O/H)ob
served 8.36
30 Dor
30 Dor
S311
S311
10
1st conclusion Yes, one can easily reproduce th
e temperatures and abundances implied by the
collisionally excited lines using simple
photoionization models.
But what about the recombination lines
of heavy elements? Can we reproduce them?
Models with 12log(O/H)in 12log(O/H)REC 8.7
,8.8 have lower Te's than required. Hence, to
produce the required Te contrast, let's introduce
extra heating (due to some unidentified process)
in localized zones. This will change the Te str
ucture...
11
First, consider the Te structure implied by the
simpler models
ne 300 cm?3 12log(O/H)in 8.5, 8.6
ne 1000 cm?3 12log(O/H)in 8.5, 8.6
recombination contribution to N II ?5754
M16
M17
M20
M8
NGC 3576
Stasinska (1980) hardening of the radiation
field with depth into the cloud O III
efficient cooling
12
ne 300 cm?3 12log(O/H)in 8.3, 8.4
ne 5000 cm?3 12log(O/H)in 8.5, 8.6
NGC 3603
M42
30 Dor
S311
13
12log(O/H)in 8.75, 12log(O/H)out 8.58,
ADF(O) 1.5
extra heating 7 of total
(See also Binette Luridiana 2000 Binette et
al. 2001)
O III emitting region
N II emitting region
Te(N)?Te(O), and O III more sensitive to
increased Te
14
Changing the width and height of the peaks of
extra heating
for models with similar characteristics to some
of the objects
M16
M20
M8
(Teff 39000 K 12log(O/H)in 8.75)?
ne 300 cm?3
ne 1000 cm?3
Introducing cold regions (metal-rich inclusions)
will have a similar effect (lower values of TeN
II/TeO III)
15

• 2nd (and last) conclusion
• Three alternatives to explain the
  observations
  
• Temperature fluctuations. One would need a
  heating mechanism whose efficiency increases with
  distance from the star, so that the final Te
  structure mimics the one found with simple
  photoionization models.
• Metal-rich inclusions. In this case, to reproduce
  the observed Te structure one could change the
  value of N/O in the inclusions (as Tsamis
  Péquignot 2005 did in their model for 30
  Doradus).
• Invoke Occam's razor and use the simpler models.
  (Errors in the recombination coefficients?)?

Mars, Polar Geology (HIRISE/MRO)?
Credit NASA/JPL/University of Arizona