Local correlations of the fluctuations at different scales of the 60 and 100 $\mu$m emissions of a high galactic latitude complex
A. Abergel, F. Boulanger, J. M. Delouis, G. Dudziak, S. Steindling;
AaA, 1996, 309, 245

ABSTRACT:The local correlations of the brightness fluctuations at different angular scales in the range 1.5-48 arcmin are analysed on a large interstellar complex lying at high galactic latitude, using one 12.5x12.5deg map at 100 and 60{mu}m of the IRAS Sky Survey Atlas (ISSA). A wavelet analysis is first applied to produce filtered images containing the local brightness fluctuations at different angular scales. These filtered images allow computation for each pixel of the amplitude of fluctuations. The local spectral indexes are then obtained for each pixel, by use of a power law to fit the variations with the scale of the amplitudes. For most of the analysed field (83% of the pixels at 100{mu}m), the amplitudes of fluctuations at different scales are correlated, and decrease with decreasing scale. The variations of the mean statistical fluctuation at 100 {mu}m with the angular scale {theta} in units of arcmin are fitted with the formula: F({theta})=0.12({theta}/1.5)^0.67^MJy/sr. Maps of the spectral indexes look significantly more homogeneous than the brightness maps, which demonstrates that the correlation between the amplitudes of fluctuations at different scales does not appear to depend on the total quantity of matter along the line of sight. The distribution of the spectral indexes is gaussian. The central value, after correction from the noise component, does not coincide with the values of the exponent of the Larson law n(l){prop.to}l^p^ found in the literature and deduced from molecular data. This difference is likely due to different origins of far infrared and molecular emissions, and to systematic biases toward high extinction regions of molecular observations. Several point-like or filament-like bright features are singular, because the amplitudes of the fluctuations at different angular scales are not correlated, which indicates a change in the physics of the cloud hierarchy from largest to smallest structures. The comparison of 60 and 100{mu}m maps shows that these features come from the coldest regions of the analysed field. These cold regions are likely molecular emitting regions, which could explain the high dispersion of densities computed for a given scale from molecular data.
KEYWORDS: methods: data analysis, techniques: image processing, ism: clouds, infrared: ism: continuum, dust
PERSOKEY:dust, STATISTICAL ANALYSIS, fir, iras, ,
CODE: abergel96a