Statistical descriptors of clouds and clusters (astronomy)

Molecular clouds and the star clusters they spawn are chaotic systems, due to the non-linearity of the processes that drive their evolution. It is therefore not normally sensible to try to simulate in detail a particular region; rather one needs statistical descriptors of the structures that are seen, which can then be used to compare one region with another, and/or with the results of simulations to ascertain whether the latter are capturing realistically the processes at work.
There are many algorithms for doing this in other fields of science. The aim will be to develop and refine these algorithms, and apply them to star-forming molecular clouds and young star clusters, so that recent advances in the power of telescopes and in the scope of numerical simulations can be more fully exploited. The student will become expert in handling continuum and point data sets, both from observations and from simulations; in designing, extracting and analyzing statistical descriptors of chaotic systems; and in tackling the inverse problems associated with constraining the threedimensional structure of an astronomical source that is seen in projection from
a single angle.