Computational intelligence techniques for submillimetre polarization modeling

Detailed modeling of submillimetre continuum and polarization data in the submillimetre involves a massive-scale parameter search to optimize fits to continuum and polarization data simultaneously. We introduce a new modeling technique based on an advanced multi-objective genetic algorithm that solves these problems very efficiently. We model the 850 mu m continuum emission of the G11.11-0.12 filamentary infrared dark cloud, and predict the corresponding polarization patterns. We also show preliminary fits to the continuum and polarized emission of a clump within the OMC-3 region of the Orion filament. Finally, we discuss how this technique might be used to model the large polarization data sets that future SCUBA-2 molecular cloud surveys would produce.