Interstellar dust in dwarf galaxies

Resumen

Instead of being a passive observer, interstellar dust plays a vital role in the evolution of galaxies. Due to its high opacity for the ultraviolet (UV) and optical photons, interstellar dust is effective in attenuating, via absorption and scattering, the UV/optical spectral energy distribution (SED) of galaxies. Besides, interstellar dust provides about a third of the total Galactic luminosity via their infrared (IR) emission. In this thesis we have studied the interstellar dust of dwarf starbursting galaxies from both the perspectives of the attenuation and the self-emission of dust grains. We have chosen these kind of objects because we are interested in understanding how the dust behaves in environments that are very strongly influenced by intense star formation (SF) processes and by the action of stellar winds coming from the massive stars. Starbursting dwarf galaxies are ideal objects to study this as they have, in general, a small number of SF regions dominating the emission. Therefore, a clear separation between the emission coming from the SF regions and the diffuse disk can be performed. In the first part of this thesis we studied the dust attenuation in the dwarf starbursting galaxy NGC4214. The UV/optical emission of NGC4214 is dominated by two SF complexes located at its centre. These regions were mapped with the Wide Field Camera 3 (WFC3) onboard the Hubble Space Telescope (HST) covering a broad range of the spectrum, from the near-ultraviolet (NUV) to the near-infrared (NIR), including the hydrogen recombination lines H_beta, H_alpha and P_beta. We made use of these hydrogen recombination lines to compute the attenuation in the central SF complexes of NGC4214. Our aim was to obtain a high resolution picture of the distribution of the dust attenuation associated with these complexes and to compare it to the distribution of the dust emission. Furthermore, the attenuation maps allowed us to derived attenuation-corrected maps of the ionised line emission maps. The attenuation was calculated assuming a foreground dust screen model. We tested two different extinctions laws described by the values of Rv=3.10 and 5.45. The extinction law of Rv=3.10, which is widely used in the literature, describes the dust properties found in the diffuse interstellar cirrus, whereas the extinction law of Rv=5.45 is more appropriated for dust in dense molecular clouds. The choice of Rv=5.45 was based on the results found by Úbeda et al. 2007. Three attenuation maps for each value of Rv were derived based on the P_beta/H_alpha, P_beta/H_beta, and H_alpha/H_beta line-ratios. If the assumed geometry (foreground screen) and the dust properties (Rv=3.10 or Rv=5.45) are correct, then all three line-ratios should give exactly the same result. Deviations from this situation allow us to draw conclusions about the real dust geometry and properties. In general, the different attenuation maps show very similar structures of the dust attenuation which furthermore agree qualitatively with those of the dust emission at 8 microns which is the only dust emission map with a comparable spatial resolution. However, the exact value of the attenuation at a position is different depending on the line-ratio considered. In order to quantitatively compare the attenuation maps obtained with different line-ratios, we compared these maps on a pixel-by-pixel basis. Our study was restricted to those regions where the photo-ionisation is the dominant process in the ionisation of the gas. We found good correlations between the different attenuation maps, but this correlation is not always linear and in some cases has a large scatter. Specifically, we found that the photo-ionised regions where we measured the highest values of the attenuation present clear deviations from linearity. In general, the attenuation maps derived with the line-ratio P_beta/H_alpha present higher values than the maps derived with P_beta/H_beta, and the latter show higher values than the map derived with H_alpha/H_beta. These trends are less pronounced but still present for the extinction law of Rv=5.45. We interpret the fact that the values of the attenuation derived from different lines agree better for Rv=5.45 as an indication that this value is more appropriate for the dust properties in NGC4214. We discuss different geometrical scenarios that could explain our other findings. The fact that the attenuation derived from P_beta/H_beta is lower than the attenuation from P_beta/H_alpha for several regions can be understood if scattering has an important effect in bringing back photons to our line of sight. Scattering is more important for the shorter-wavelength H_beta-line than for H_alpha and thus produces a lower value for P_beta/H_beta than for P_beta/H_alpha. Scattering is becoming relevant if the dust is close to the emitting gas. A simple mixing of dust and gas is not able to reproduce the high values of the attenuation found in the SF regions of NGC4214. Finally, the fact that the attenuation derived from H_alpha/H_beta is lower than the attenuation from P_beta/H_alpha is an indication for a clumpy distribution of the dust. All these indications lead to a final picture in which the dust causing the attenuation is mainly distributed in dense, clumpy clouds close to the emitting region, which physically would correspond to dust within a fragmented photo-dissociation region (PDR). In the second part of this thesis we analysed the interstellar dust of NGC4214 and NGC4449 from the perspective of its re-emission at far-infrared (FIR) and submillimetre (submm) wavelengths. The goals were to test whether models with Galactic dust properties could describe the dust SED of these dwarf galaxies, to test the relevance of the geometrical distribution of dust and stars on the dust SED and to carry out a detailed energy balance of the dust heating and emission in a realistic geometry. Due to their proximity we were able to derive the dust SED separately for the emission from the major massive SF regions and for the diffuse dust. For the first time this analysis was done from the perspective of a self-consistent radiation transfer calculation constrained by the observed SEDs of spatially separated components on resolved maps. In making predictions for the UV/optical-FIR/submm SEDs of these components this analysis quantitatively takes into account the level and colour of UV/optical radiation fields incident on both pc-sized dusty structures in the SF complexes and on dust grains distributed on kpc scales in the diffuse ISM, illuminated by a combination of UV light escaping from the SF regions and the ambient optical radiation fields, as constrained by UV/optical photometry of the galaxy. The overall analysis was done using the Popescu et al. (2011) model based on full radiative transfer calculations of the propagation of starlight in disk galaxies adopting a realistic geometry for the distribution of stars and dust. For the description of the dust emission from HII regions and their associated PDRs we used the model of Groves et al. (2008). This model describes the luminosity evolution of a star cluster of mass Mcl, and incorporates the expansion of the HII regions and PDRs due to the mechanical energy input of stars and supernovae (SNe). The dust emission from the HII region and the surrounding PDR is calculated from radiation transfer. The large amount of ancillary data and results from previous studies allowed us to constrain a major part of the input parameters of the models. We achieved a good agreement between data and models, both for the diffuse dust emission and the dust in HII+PDR regions for both galaxies. We achieved satisfactory fits for the star forming regions with the exception of the IRAC 8 microns data points. Possible reasons for this discrepancy are that the model assumptions on polycyclic aromatic hydrocarbons (PAHs) abundance and destruction are not completely adequate for the case of NGC4214 and NGC4449, or that these galaxies have an unusually high emission at 8 microns for their metallicities and radiation field, which, in the case of NGC4214, is supported by other studies (Engelbracht et al. 2008). For both galaxies, we could fit the diffuse dust SED satisfactorily, and we derived global gas-to-dust mass ratios compatible with their expected values from their metallicities. However, we inferred that the attenuated UV emission predicted by the modelling is a factor of 2-4 lower than the observed, diffuse UV flux. We discussed different explanations for this discrepancy (escape of UV emission, geometrical effects, a very extended dust disk and different dust properties). The most plausible one is that a large fraction of the UV radiation that escapes from HII+PDR regions leaves the galaxy unattenuated and is thus not participating in the heating of the diffuse dust, most likely due to a porous ISM. For the first time an analysis of the SED of dwarf galaxies based on radiation transfer models and separating the emission from the SF regions and the diffuse component has been performed. The methodology applied here will be used in the future to analyse a sample of dwarf galaxies with different inclinations and at different evolutionary states. This will help us to understand better the dust properties and dust heating mechanisms in this particular type of galaxies. The conclusions will have important consequences in quantify the star formation rate in this type of objects.