Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.atnf.csiro.au/research/LVmeeting/magsys_pres/ISM_SMC.pdf Дата изменения: Sun Aug 5 03:59:00 2007 Дата индексирования: Tue Oct 2 17:18:06 2012 Кодировка: Поисковые слова: п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п п

Zooming into the ISM of the SMC
1

Erik Muller1, Paul Jones2 Bolton Fellow: ATNF,CSIRO 2 UNSW


Preamble:
· · · · The Turbulent ISM A description of turbulence: its role and importance Observations of the SMC & technique Caveats & future work

Erik Muller, E+FW conf, July 2007


The turbulent Interstellar Medium
· The ISM is best traced with Neutral
hydrogen (HI)

· Energy injection/distribution
­ reshapes ISM at all scales.

· Mixing of the ISM
­ distribution/enrichment of metals

· Influences SFE
­ Turbulent support against core collapse
Erik Muller, E+FW conf, July 2007


Characterising Turbulence
The ISM is almost ALWAYS turbulent
­ Large scales, low viscosity ­ Reynolds numbers are typically ~1000 ­ Incompressible fluid flow: P(k)=k k ~ -11/3 (3-D kolmogorov index)

Erik Muller, E+FW conf, July 2007


Characterising Turbulence
· Fractal nature of the ISM:
­ Outer scale: The largest scale at which energy transfer occurs:
· Spiral arms, Colliding systems

­ Inner scale: The smallest scale at which damping is significant ­ Reynolds number becomes ~1
· ~molecular level (dept. on T, P etc.).

· Energetic processes re-organise the ISM ­ eg. SNe
Erik Muller, E+FW conf, July 2007


The effects of Starformation/SNe
· · · · Early stages: core clumping of the ISM. Stellar winds ­ sweeping/heating the ISM PNe, SNe, shells and shell fragmentation Scales ~20pc.

Ryan Joung & Mac Low, 2006 Simulated SN-driven turbulence in a compressible ISM. No specific `scale' 90% of EKin at L<180pc

(Ryan Joung & Mac Low, 2006)
Erik Muller, E+FW conf, July 2007


Why the SMC?
· · · · · · · Far enough - ~Parallel lines of sight Close enough ­ high spatial resolution Active starforming region in the SW SMC Among largest known HI column densities
­ ~1022 cm-2.

Augment with existing HI data:
­ Stanimirovic: ATCA+Parkes

Accessible resolution: ~2-3 pc Problems ­ Possible large line of sight depth. ­ Optical depth.
Stanimirovic: http://www.atnf.csiro.au/research/smc_h1/

Erik Muller, E+FW conf, July 2007


The Energetic Southwest SMC.

Courtesy: Karl Gordon

Muller & Jones, In prep
Erik Muller, E+FW conf, July 2007


Target observations.
· · · · Three overlapping positions 7x12 hr @ 6B, 2x12 hr # 750A Beamsize ~5 arcsec @ 3K/chan (6.7 km/s) Missing short spacings not necessary

Erik Muller, E+FW conf, July 2007


Total power

Erik Muller, E+FW conf, July 2007


SMC Spatial power spectrum
· SPS from entire of SMC
­ ~ -3.3 Kolmogorov
· Incompressible index?

­ Featureless
· No structure-dependent processes

~ -3.05

­ Max Limited to ~60 arcsec (17pc)

After Stanimirovic, 2000 Erik Muller, E+FW conf, July 2007


Calculating the SPS
Input ISM dataset Multiply Data with `Apodizing' mask Create tapered dataset Create zero-padded, blank (=1) dataset (same dimensions of ISM dataset) Dummy gaussian, psf~20 pixels

FFT Real Imaginary Azimuthal median. Equal intervals in log.
Erik Muller, E+FW conf, July 2007

P=I2+R2

Fit to obtain




Early, Hi-resolution Power Spectrum
144 km/s
Typical Error

Fitted component: -3.02
Outer-radius of sample

Beam

Interferometer data only

Erik Muller, E+FW conf, July 2007


Coping with optical depth.
· Optical depth presents a significant challenge:
­ Absorption appears to be more important at small scales. (e.g. Gibson et al, 2000). ­ Artificial steepening of the SPS at small scales.

Peak HI.

Erik Muller, E+FW conf, July 2007


The extended SMC
Simulated Radial distribution of HI in SMC

~25 kpc
· Uncorrelated, random velocity fluctuations · Disentangle using `Velocity component' Technique · averaging over large velocity range · 3D Density Distribution.

Muller & Bekki, Submitted

Erik Muller, E+FW conf, July 2007


Summary
· Highest-yet resolution observations of the SMC.
(and the highest possible, until SKA)

· Powerspectrum index of (~-3) appears to extend into high frequencies · Not possible to exclude excess power at <10pc · Refinement of process still underway.
Erik Muller, E+FW conf, July 2007