[tel-00726959, v1] Caractériser le milieu interstellaire ... - HAL - INRIA

H. S. Liszt and J. Pety: Imaging diffuse clouds: bright and dark gas mapped in CO51˚20'f-19,-16e13.3f-16,-12e24.6f-12,-9.5e13.5-9.5,-7ef9.7-6,-1.5ef15.9allef36Declination(J2000)51˚10'51˚00'50˚50'50˚40'gab4h00mpkcd0.43h58mgab4h00mpkcd0.43h58mgab4h00mpkcd0.43h58mgab4h00mpkcd0.43h58mgab4h00mpkcd0.43h58mgab4h00mpkcd0.43h58mRight Ascension(J2000)tel-00726959, version 1 - 31 Aug 2012T r* [Kelvin]40 g30 f20 e10 d0 B0355-10 c-20 b-30 a-40 pk-20 -10 0V LSR (KM S -1 )T r* [Kelvin]e -τ -10 -1 0355+508/32 HNC1 CO0 HCO +-1 H I-2-40 -20 0V LSR (KM S -1 )Fig. 12. The sky field around the position of B0355+508. At top are maps of integrated CO intensity made over the velocity intervals indicatedin each panel, corresponding to the five strong components of the HCO + absorption profile seen at lower right. CO emission profiles at variouslocations indicated in the map panels are shown at lower left. CO emission profiles toward B0355+508 and averaged over the map area are shownabove the absorption line profiles.(the usual last contour on CO sky maps) are detected at orabove the 90% confidence level. To put these brightness andsensitivity levels in context, note that there is a straightforwardrelationship between W CO , f H2 ,andE B−V once the CO-H 2and E B−V /N(H) conversion factors are fixed; for the standardW CO /N(H 2 ) = 2 × 10 20 cm −2 H 2 (km s −1 ) −1 and N(H)/E B−V =5.8 × 10 21 cm −2 mag −1 one has W CO = 14.5 f H2 E B−V Kkms −1 .At E B−V = 0.1 mag, emission only slightly exceeding 1 K km s −1implies a molecular fraction f H2 > 1 and therefore is too brightto be accomodated by a CO-H 2 conversion factor as large as thestandard 2 × 10 20 cm −2 H 2 (km s −1 ) −1 .Shown in each panel of Fig. 15 are lines representing theCO emission expected if various fractions f H2 of the total neutralgas column are in H 2 with a typical galactic W CO /N(H 2 ) conversionfactor X CO = 2 × 10 20 H 2 cm −2 (K km s −1 ) −1 .Muchof the CO in Fig. 15 occurs above the line corresponding tof H2 = 1 and is therefore too bright to be accomodated by theusual CO-H 2 conversion factor; indeed, almost every CO linewith W CO> ∼ 1Kkms −1 may be described as overly-bright inthis way if f H2 = 0.5, hence the great majority of all the statisticallysignificant emission represented in Fig. 15 andinthemaps shown earlier for these sources. For the brightest pixelsN(H 2 )/W CO < 5 × 10 19 H 2 cm −2 (K km s −1 ) −1 .The same W CO -E B−V diagrams are shown for sources withhigher E B−V in Fig. 16. Much of the gas around B2200+420falls above the line for f H2 = 1, and attains such high brightnessthat its H 2 /W CO ratio is 3−4 times below the standard conversionfactor. However, this case becomes increasingly harder to maketoward the other sources having higher E B−V as in the bottompanels of Fig. 16.8.2. Sub-structure in reddening would not eliminate largeW CO /E B−V ratiosCO emission is heavily structured on arcminute scales, well belowthe 6 ′ angular resolution of the reddening maps, and the highvalues and large scatter in W CO /E B−V in Fig. 15 cannot be accomodatedwith a fixed ratio of W CO /N(H 2 )orW CO /N(H) exceptby positing strong unresolved variations, essentially clumping,in E B−V . It is important to understand the extent to whichthis might represent unresolved structure in the total columndensity, for instance with regard to cleaning maps of the cosmicmicrowave background (Planck Collaboration 2011). Giventhe extreme sensitivities of the CO abundance and brightnessto N(H 2 )indiffuse clouds and the fact that even f H2 may varyin diffuse material, it is entirely possible that the large contrastsseen in W CO do not have strong consequences for the distributionof N(H), E B−V ,orevenN(H 2 ).Shown in Fig. 17 are cumulative distribution functions ofthe integrated CO emission W CO in the fields around B0954+658A58, page 15 of 23