Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.sao.ru/hq/katya/NDG/DVD/Talks/Roychowdhury.pdf Äàòà èçìåíåíèÿ: Wed Feb 24 16:36:22 2010 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 15:38:40 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http www.astronomy.ru forum index.php topic 4644.0.html

Star Formation in Extremely Faint Local Volume Dwarf Galaxies
Sambit Roychowdhury with Jayaram N. Chengalur Ayesha Begum, Univ of W-Madison Igor D. Karachentsev, SAO, Russia September 14 , 2009


Faint Irregular Galaxy GMRT Survey (FIGGS)

· Galaxies identified using the selection criteria: MB > -14.5, HI Flux > 1 Jy km s-1, Dopt > 1', > -40 · By far the largest study of HI in dwarf galaxies · Total of 62 galaxies obser ved


The FI G G S ga l a x i e s


Galaxy Formation

· In heirarchical galaxy formation models small objects collapse first, and in turn merge to form large galaxies

· Simulations accurately trace only collapse and merger of dark matter hal os

· Bar yon physics, viz. accretion into these halos, star formation and feedback are poorly understood

· Simulation of galaxy formation necessarily use star formation 'recipes' These recipes are derived from obser vations of nearby galaxies


Recipe one : Schmidt's Law · Schmidt in 1959 proposed: S F R=A N s ga · Kennicutt (1989, 1998) used H, HI and CO obser vations to get a Law for (optical) disk averaged values: (2.5±0.7) â 10-4
1.4±0.15 gas M y r-1 kpc-2 -2 1 M pc


Recipe two : Star Formation Thresholds · Toomre in 1964 first suggested the idea of a threshold, related to a critical density for gravitational instability in a thin rotating gaseous disk · Skillman in 1987 proposed that star formation occurs only above a threshold column density (1021 atoms cm-2) for a resolution of 500 pc, which may be related to a critical amount of dust sheilding required for molecular gas formation · Kennicutt (1989), Mar tin & Kennicutt (2001) show SFR as traced by H show clear thresholds (nearby spiral galaxies)


Star Formation Thresholds · Kennicutt in 1989 combined obser ved H distributions with HI and CO data to find the threshold column density · These thresholds represent the region where the disk goes from being dominated by molecular gas to being dominated by atomic gas


Star Formation Recipes for dwarf galaxies

· All studies till now mainly looked at normal spirals or starburst galaxies

· Why should we look at dwarf galaxies? Dwarf galaxies impor tant because they lie at the beginning of the heirarchical galaxy formation scale Also provide unique laboratories, low metallicity, low dust content, low pressure, etc.

· We will concentrate on a sample of local volume dwarf galaxies


DATA

· HI data taken from Faint Irregular Galaxies GMRT Sur vey (FIGGS), median MB -13.0, median mass 3 â107 M (Begum et al., 2008)

· UV data obtained from GALEX public releases, last release in June 2008 circular images of the sky with 1.2 diameter and 5 resolution ° ° FUV band (1350 A to 1750 A) data used

· From the FIGGS sample we could find 23 galaxies which had GALEX data, and for which we could reach linear resolutions of 400 pc


The Sample


The Sample


DATA ANALYSIS

· Tasks in Classic AIPS used to reduce data

· For HI data, data cubes of resolution 400 pc were made from available continuum subtracted data using IMAGR

· Background subtracted FUV maps were aligned to their corresponding moment 0 maps using HGEOM, then smoothed to required resolution using SMOTH (or CONVL)


DATA ANALYSIS


DATA ANALYSIS

· FUV data conver ted to SFR using mGALE X = - 2.5 log(cps) mAB = mGALE X + 18.82 S F R(M y ear-1) = 1.4 â 10-28L (erg s s-1 H z -1) ° (from Kennicutt 1998a, valid within the range 1250-2500 A)

· HI data conver ted to column density using NH = 1.8224 â 1018
-

Tb( )[K ] dv [K m s-1]

kT where 22 b = B


for Disk Averaged Values

· The task BLSUM was used to extract the total flux density over the optical disk from the respective data cubes, which was then conver ted to column d e n s i ty

· The task BLSUM wa images coming from tical disk, but having edge. This SFR is for the four galaxies available.

s used to get the average FUV flux in the smoothed the elliptical region having ellipticity same as the opSFR equal to 1.85 â 10-4My r-1kpc-2 at it's the average UV flux at the edge of the optical disk in our sample for which only Holmberg radius was


Global Kennicutt-Schmidt law


Kennicutt-Schmidt law for spirals and dwarfs


for Pixel by Pixel Comparison · Moment 0 maps constructed using MOMNT · Data over each 'pixel' (as shown below) averaged over

· For 10 galaxies, a similar procedure as described before was followed to get pixel by pixel comparisons at linear resolutions of 200 pc


Pixel by Pixel Comparison


Pixel by Pixel Comparison


Pixel by Pixel Comparison


P by P C


Simulation


Summary · The globally averaged gas density in our sample galaxies lies below most estimates of the star formation rate, and the obser ved SFR is also lower than that estimated from the Kennicutt-Schmidt relation · The data is better fit by the steeper slope found for the spirals only sample by Kennicutt (1998) · For most (18/23) galaxies SFR can be parametrized to have a power law dependence on gas when both parameters are measured on 400 pc scales, however the coefficient and index of the power law varies substantially from galaxy to galaxy · The index of the power law is in general steeper than the value of 1.4 for the Kennicutt (1998) relation. The obser ved SFR rate is in general lower


than that predicted by this relation, with the discrepancy decreasing at the h i g h e s t g a s c o l u m n d e n s i ti e s

· At 400 pc resolution the SFR continues to fall smoothly until one reaches the sensitivity limit of our obser vations, and there is no evidence for a "threshold density" below which star formation is completely cut off

· At 200 a n d th e a n d th e th e g a s

pc resolution, offsets between the sites of current star formation locations where the gas density peaks become more pronounced, SFR can be parametrized as having a power law dependence on density for only 5/10 galaxies

· For the majority of these galaxies (4/5) the power law index measured at 200 pc is flatter than that measured at 400 pc


Caveats

· Assumptions: Ver y little molecular gas present The stars have solar metallicity and Salpeter IMF The galaxy has had continuous star formation over time-scales of 108 years or longer Internal and external dust attenuation has been correctly accounted for


Caveats · Effect of taking molecular gas into account shown by brown arrow · Effect of taking sub-solar metallicities shown by blue arrow


Caveats · Truncating the IMF at the high mass end, presence of unaccounted for dust, or if the galaxy has not undergone star formation in the recent past, would affect the results as shown by the blue arrow · Brown arrow shows how the results will change if the star formation is dominated by a recent star burst


Caveats
1

0.8

0.6 FUV-NUV 0.4 0.2 0 0 10 20 30 MHI (10 M)
6

40

50

Disc averaged FUV-NUV colour of 22 galaxies, plotted against mass of HI gas contained within optical discs, having an average value of 0.34


Caveats

· According to Boissier et al. (2008), FUV-NUV color of 0.34 can result from the following models: Galaxy has a Kroupa IMF with metallicity = Z/20, an age of 6.8â108 years, and star formation was quenched after 108 years Has a Kroupa IMF truncated at 5M at the highend, solar metallicity, has been forming stars for 4.3â108 years continuously Has an age of 2.4â108 years and star formation was quenched after 108 years Has Kroupa IMF, solar metallicity, an age of 2.4â108 years and star formation was quenched after 108 years


Caveats

· Bigiel et al. (2008) did a similar study with THINGS galaxies, includes dwarfs, but pixel by pixel correlation not done for the four galaxies below t h e FI G G S m a g n i t u d e l i m i t For the 4 irregular galaxies for which they do fit a power law, they find power law indices that var y from 1.59 to 2.78 They also find that the "star formation efficiency" (i.e. SFR/gas ) is lower in dwarfs and the outer par ts of spirals than in the inner, H2 dominated regions of spiral galaxies They also find that the gas density in dwarfs truncates shar ply at about 9M/pc2 (study done at 750 pc linear resolution)


Caveats
0.14

fraction of combined mass of all galaxies

0.12

0.1

0.08

0.06

0.04

0.02

0 -1.5

-1

-0.5

0

0.5 Log HI (M pc )
-2

1

1.5

2

2.5

Histograms depicting what fraction of combined mass of all galaxies lie in a p a r ti c u l a r c o l u m n d e n s i ty b i n Appears that HI gas occurs in small clumps whose density gets smoothed out when one obser ves with coarser resolution


MNRAS 397 1435-1453

Thank You