Southern Supershells

1
Southern Supershells

• Naomi M. McClure-Griffiths
• University of Minnesota
• HI Surveys Workshop
• May 21-23, 2001
• NRAO, Greenbank

2
Outline

• Introduction
• Background on HI Shells
• New SGPS shells
• HI supershells
• Chimneys GSH 2770036 GSH 2800059
• Coalsack shells GSH 304-00-12 GSH 30501-24
• GSH 299-0073, GSH 297-0074, GSH 293-0254
• Galactic distribution of HI shells
• Distinguishing shells from random fluctuations in
  the ISM
• Future work on SGPS Shells

3
HI Shells Background

• Among the largest objects in the ISM
• Sizes vary from tens of parsecs to kiloparsecs
• HI shells are usually seen as voids in the
  Galactic HI
• Supershell masses range from 105 to 106 M?
• Expansion (formation) energy (Chevalier 1974)
• EE 5.3x1043 no1.12 R3.12 v1.4
• Typical formation energies are on the order of
  1051-1052 ergs
• Formed by stellar winds and/or supernovae
• Largest of shells, with energies in excess of
  1053 ergs, may be caused by gamma ray bursts or
  HVC impacts
• If the shell exceeds the scale height of the
  Galaxy it may form a Galactic chimney,
  supplying hot gas to the Galactic halo

4
Supershell Identification
5
Supershell Identification
front wall
rear wall
6
Galactic Supershells GSH 277036

• Large void detected in the HI at l277º, b0º,
  v36 km/s
• Previously identified as an interarm void
  adjacent to the Carina arm
• Distance 6.5 kpc
• Physical scale
• Radius 305 pc
• Height gt 1.1 kpc
• vexp 20 3 km/s
• Mswept 2.7 5.6 x 106 M?

McClure-Griffiths et al. 2000
7
Galactic Supershells 280059

• Also extends far above and below plane (gt1.2 kpc)
• Lies on the far edge of the Carina arm
• Kinematic distance 9 kpc
• Physical parameters
• Radius 215 pc
• Mswept 1.1 x 106 M?
• Vexp 12 km/s
• EE 2 x 1052 ergs

8
GSH 277036
9
Shell formation

• Energy requirements for GSH 277036 are
  significant, so we considered various formation
  methods
• HVC impact
• Large energy deposition
• Not reliant on stellar densities
• Morphology not consistent
• Formation at the edge of the Carina arm
• Exaggerated expansion into less dense area
• Decreases energy requirements
• Widens and dramatizes the interarm region

10
Coalsack Shells GSH 304-00-12

• Rsh 220 pc
• Vexp 12 km s-1

McClure-Griffiths et al. 2001
11
Coalsack Shells GSH 30501-24

• Rsh120 pc
• vexp 9 km s-1
• Associated with Cen OB1

12
GSH 30501-24 and Cen OB1

• Use Weaver et al. (1977) wind luminosity model
  for the bubble
• Model the input wind luminosity based on 21 known
  stars
• L ½ dM/dt v?2
• For each star, we estimate dM/dt from empirical
  mass loss tables (de Jager et al. 1988) and v?
  from a stellar wind model (Leitherer et al. 1992)

13
New HI Supershells

• Three new, possibly associated shells
• GSH 299-0073
• D 15.7 kpc
• Rgal 13.8 kpc
• Rsh 315 pc
• vexp 19 km/s
• EE 3 x 1053 ergs
• GSH 297-0074
• D 14.9 kpc
• Rgal 13.4 kpc
• Rsh 355 pc
• vexp 23 km/s
• EE 5 x 1053 ergs

14
New HI Supershells

• GSH 293-0254
• D 11.5 kpc
• Rgal 11.4 kpc
• Rsh 450 pc
• vexp 24 km/s
• EE 1 x 1054 ergs
• We have found an additional 15 new shells

15
Distribution of HI Shells

• Most of the new shells are in interarm regions
• Why? Two possible reasons
• Shells correlated with star formation
• Pressure gradient on the back side of spiral arms

16
Distribution of HI Shells

• Shells are correlated with massive star
  formation, so we initially expect to find them
  where the stars are, i.e. in the arms
• Given theories of star formation and spiral
  structure do we expect to see them on the
  trailing edges of arms?
• Time for a gas to move out of spiral arm at Rgal
  8 kpc is 107 yrs
• Lifetime of a shell few x 107 yrs
• So, perhaps the shells may end up in the interarm
  regions
• Is it also possible that the pressure gradient at
  the back edge of the arms leads to a break-out
  into the interarm region that is similar to
  break-out above the plane?

17
Shells vs. Random Fluctuations

• Measurements of the HI spatial power spectrum
  show a power law from few hundred parsecs to
  parsec scales
• Goldman (2000) and Wada et al. (1999) suggest
  that the largest eddies may be kiloparsecs,
  possibly powered by differential rotation
• So, there may be stochastic fluctuations in the
  ISM on large scales that are different from the
  deterministic structures
• Can we distinguish true shells from random
  fluctuations?
• Preliminary evidence suggests that the shell
  walls may hold the answer
• Shells, when observed with high resolution, have
  thin swept-up walls random structures should
  not have these walls

18
Conclusions

• Were finding many new shells in the SGPS
• Among the most interesting are the outer Galaxy
  chimneys and the Coalsack shells
• Coalsack shell GSH 30501-24 well-modeled by the
  stars in Cen OB1
• It appears that HI shells may be preferentially
  found between spiral arms
• High resolution studies of these may allow us to
  probe the development of instabilities and the
  break-out process

19
SGPS web site
http//www.astro.umn.edu/naomi/sgps.html
20
Distribution of HI Shells

• Do we see a different population of shells
  exterior to the co-rotation radius than interior?
• Interior to Rco 9 kpc we expect star formation
  on the inside of the spiral arms
• Exterior to Rco we expect it on the outside of
  the arms
• So, can we expect large shells on the opposite
  sides?