Temp

From Tauwiki

The different FUSE bands are all highly correlated with slopes of between 0.6 and 1 (Table 1). In particular, those bands of similar wavelengths are entirely in agreement with each other. The LMC has also been observed by UIT (Parker et al.) and we have found the correlation with the UIT fluxes.

[edit] Correlation of FUV Diffuse emission with NHI

The primary source of diffuse FUV emission is the light of the hot OB stars scattered off of interstellar dust. Several works have attempted to study the connection between the FUV diffuse radiation and neutral hydrogen column density, N(HI). Understanding of this connection is important because it shows that the FUV diffuse light is due to scattering of starlight by dust well mixed with neutral hydrogen. Paresce et al (1980), Hurwitz et al (1991) and Schiminovich et al (2001) have shown existence of significant correlation between N(HI) and dust scattered diffuse UV emission but degree of correlation varies for different parts of the universe due to regional variation in properties of dust and the distribution of N(HI). We have used high resolution N(HI) observations of Magellanic clouds from Kim et al (1998) to measure the neutral hydrogen column density and figure(2) shows some correlation between N(HI) and FUSE observations upto NHI~ 2X10²¹ and beyond this, it is completely scattered. This could be due to the chaotic distribution of NHI in LMC with hundreds of clumps, shells, filaments (Kim et al 2003).

[edit] Correlation of FUV Diffuse emission with IR emission

After absorption of UV flux emanating from young OB stars, interstellar dust emits primarily in the infrared (IR) wavelengths. A relation between these two wavelengths provides valuable information about star formation rates as well as dust properties. Emission in different IR bands is attributed to different size distribution of dust. While the mid-IR emissions around 8 micron is attributed to large molecules (PAHs), the 24 micron continuum results from small grains. Emission at longer wavelengths is mostly due to large grain particles. We looked at correlations between FUV diffuse radiation and IR emission in several bands that include the four IRAS bands and 8, 24 and 70 micron SPITZER bands. Here we present the correlation of diffuse FUV emission in 1B1 band with the four IRAS bands (Fig. 3). We found a good correlation between them except for few locations in SN1987A and N11 regions. For these locations the IR flux values remain constant where as the FUV flux value shows a significant variation. This could be due to the reflection UV light from dust. We have not included one data point from 30-Doradus region and three points from SN1987A region in these correlations as these were very discrepant with respect to the average trend.

[edit] Estimation of the far ultraviolet diffuse radiation

Apart from the FUSE data , we have extensively used the UIT data to estimate the fractional contribution of FUV diffuse radiation in LMC. Wide field far ultraviolet images of the regions of nebulosity catalogued by Davies, Elliot and Meaburn (1976) has been obtained by UIT. These images includes 16 fields from LMC and 3 fields in SMC and each image is of 37' in diameter. All the images we have considered for our purpose are taken with FUV B-Camera filters: B1 () and B5 (). Parker et al (1998) have measured the total integrated aperture flux as well as stellar flux within each of the regions and also they have provided catalog of FUV magnitudes derived from point spread function photometry for the stars of these regions.

We have taken the calibrated and geometrically corrected images of UIT (2048 X 2048) which contain some of our observations and measured the flux for our observed locations. The FUSE LWRS is 30" by 30" square which is much larger than the 1.13" pixel of UIT, therefore we integrate the UIT image over 27 x 27 pixel box to compare the values. Of the 81 FUSE observations, 33 are observed by the UIT B1 filter and 29 in the B5 filter. We find an excellent correlation between the FUSE and UIT flux values as shown in fig(4). The FUSE-UIT correlation coefficients are 0.89 and 0.93 for the UIT B1 and B5 filters respectively. We determined the slope and intercept of the correlation for both the bands separately using best fit method. In case of B1 filter the slope is 2.47 and offset is -16625.2 in CU and in case of B5 filter slope is 2.18 and offset is -16096.8 in CU.

LMC shows large regional variations in morphology and composition due to vigorous star forming activity and presence of a wide range of interesting astronomical objects. There is substantial variation in the properties of dust present in different regions of LMC (Paradis et al., 2009) that results in variation in the diffuse emission from different regions. In order to assess the distribution of FUV diffuse emission in the LMC, we have estimated the FUV diffuse emission separately for different regions. The parameters involved in the estimation of FUV diffuse emission are presented in Table 1. Parker et al. (1998) have published the ratio of total stellar flux to the total integrated aperture flux for the FUV images. We have used them for estimating the FUV diffuse emission coming out from different regions in FUSE bands. The estimation is based on the following formula:

Total FUV diffuse emission(FUSE 1B1) = (Total integrated flux from a UIT region) X (Fractional diffuse flux of UIT) X (ratio of FUSE to UIT flux) (Column 6 in Table 1).

The ratio of FUSE to UIT flux for different regions of LMC has been calculated separately and fractional diffuse flux is the stellar flux subtracted total integrated flux. The calculation of unreddened stellar flux for all the FUSE bands is based on Kurucz model (Kurucz 1992). This follows the calculation of total unreddened stellar flux, i.e., sum of the contributions of fluxes from all stars within each UIT field in 1B1 band (Table 1). Input parameters for the Kurucz model include FUV magnitude of the stars contained in each UIT image ( Parker's catalog 1998) and the corresponding spectral types (Parker et al.1996). Based on this, the FUV diffuse emission at 1117 from different UIT field of LMC is estimated. The emission shows variations owing to varying stellar populations and dust distribution in different regions of LMC. We estimate a maximum of 80% FUV diffuse emission escaping from N70 region and a minimum of 17% the SN1987 region. A closer look at Table 1 shows that the stellar and diffuse flux at 1117 (FUSE band) are higher than that of UIT bands. This is obvious as the stellar population considered here is O/B type young stars that emit mostly in the FUV and also the scattering by dust increases at lower wavelengths. Variations in the fraction of FUV diffuse radiation from different regions of LMC (Table 1) provides crucial insight to the varying dust population in these regions and also supports the idea of dominating local effects at FUV wavelengths.

Different regions in LMC have significantly varying stellar populations and dust and gas distribution. While 30 Dor, N11, N4, N70 regions are rich in OB associations and young stars, the LMC bar region is mostly populated by old stars. These variations significantly affect the FUV radiation field and in turn the FUV diffuse background radiation. A similar variation exists in the dust composition in various LMC regions. The LMC bar region is rich in gas and molecules incorporated with fresh dust out flowing from AGB stars, 30 Dor and the regions around it have a more evolved and relatively bigger size dust population (Paradis et al. 2009). A more detailed description of the variation of FUV diffuse emission with varying dust population in different regions of LMC will be the focus of a future manuscript.

The median value of the fractional FUV diffuse radiation escaping LMC (column 8 in Table 1) comes out to be 38%. This result predicts that a significant fraction of stellar light escapes in the form of diffuse light scattered from dust. This is important in the context of the fraction of ionizing radiation escaping galaxies (Leitherer et al.1995) that keeps the IGM ionized. At lower wavelengths, the fraction of diffuse light should constitute even higher fraction of the total flux escaping LMC. Since, the optical depth at 900 is even higher, much of the light escaping at Lyman continuum wavelength should be dust scattered star light. The results reported here become even more important as the escape fraction of ionizing photons depends on the topology of the ISM (Heckman et al.2001).