Gareth Jones

This poster presents ongoing work exploring prescriptions for dust re-emission models which are self-consistent with the stellar population, generated using Binary Population and Spectral Synthesis (BPASS) models. Preliminary girds of dust models have been generated using the da Cunha et al. (2008) energy balance prescription and tested by fitting to stacked photometric data from the COSMOS-2015 catalogue. Fits to stacked galaxy samples are shown within Figures 1 and 2. In Figure 1, it can be seen that the model is matching closely the COSMOS observational data in both the visible and far-infrared parts of the spectrum, and so both stellar and dust parameters are being constrained. The IRAC filter data have not been fitted as a further component is still required in these preliminary models to match observations here. However, when looking at Figure 2, it can be seen that while the stellar component is still being well constrained, the dust component is not as the model is not generating enough flux in the infrared regime to match the observational data. This shows that while our model is working for some observational galaxies, more still needs to be included within other components (such as a highly extinct population) to get a better match between our model and observations.

Therefore, we have looked into using already available fitting codes and have been testing BAGPIPES which implements Draine & Li (2007) dust models. The same galaxy group shown in Figure 2 is then also fit using BAGPIPES and shown in Figure 3. Now it can be seen that while the model is still able to fit to observations in the optical part of the spectrum, the model is now also fitting the infrared data as well, achieving the same flux levels, although it does not have the full same shape as the observational points, something that will be investigated by looking into alternate dust prescriptions. This better fit is most likely due to the fact that BAGPIPES fits the stellar and dust parameters simultaneously, causing it to pick a population which contains enough dust to also constrain the infrared data.

Within Figure 4, the dust parameters of the Draine & Li (2007) model are varied to show that each observational data point for the galaxy composite in Figure 3 can be constrained by at least one model. This is done by varying one of the dust parameters constraining the Draine & Li (2007) models at one time, while holding the others fixed. In the top panel, increasing the lower limit of the starlight intensity distribution (U_min) causes the far-infrared emission to shift to hotter blackbody emission temperatures, and so a peak emission at shorter wavelengths. The middle panel has increasing fractions of stars at U_min (gamma) causing the mid-infrared emission to increase. The bottom panel has increasing PAH mass fractions which causes the emission in PAH at the shortest infrared wavelengths to increase. While most infrared data points are fairly constrained by the original best fit dust model, a large decrease in the gamma parameter value would allow for the MIPS 24 micron data point to be constrained.

Figure 5 has a comparison of the stellar parameter space determined for each stacked galaxy spectrum using BAGPIPES when fitting with BC03 and BPASS stellar models. The resulting space has BPASS models tending to be older populations than the BC03 models, and slightly more massive, with no characteristic differences in the extinction levels of the population. This emphasises the systematic uncertainties that arise when relying on a single model prescription and its implicit assumptions.