Skip to main content

Onset of Star Formation: Connecting Theory and Observations

Final Report Summary - SFONSET (Onset of Star Formation: Connecting Theory and Observations)

The formulation and testing of a star formation theory is essential in our quest to understand the origin of stars and planets, including that of our own Sun, of our solar system, of our planet, and, ultimately, of life itself. However, the mechanisms that are responsible for initiating and regulating star formation remain under debate, even after many decades of theoretical and observational studies. Molecular clouds, the birthplaces of stars and planets, are comprised of many different interacting components, including neutral and ionized gas, dust grains, and magnetic fields. Gravity, ionizing processes such as cosmic radiation and high-energy photons from massive stars, and supersonic turbulent motions affect the dynamics of molecular clouds, while an intricate network of astrochemical reactions determine the molecular composition and the charge distribution throughout the cloud.

Although molecular clouds can have very high masses (up to a million solar masses), they appear to be globally supported against gravity, while a small fraction of their mass forms high-density fragments, which then go on to collapse into stars. In this way, the efficiency of star formation is very low. Since molecular clouds are observed to be very cold, thermal pressure forces are unable to account for their global support against gravity; rather, clouds must be supported by a combination of effective pressure from magnetic fields and supersonic turbulence, although the relative importance of each is still fiercely debated.

The reason for the large number of remaining open questions in the field is that observable quantities, such as the continuum emission from interstellar dust, or maps of molecular line emission, are typically affected by multiple factors, which are very hard to deconvolve. This results to degeneracies between predictions of different models that are difficult to break. The aim of project SFOnset was to rectify this problem, by creating chemodynamical models that follow all processes affecting interstellar gas probes self-consistently and simultaneously. The ultimate goal has been to identify diagnostic tests that minimize degeneracies between different models, and, as a result, have maximum potential for discriminating between different theories of star formation.

The SFOnset project has resulted in a total of 18 publications in refereed journal, and the training of three postgraduate students in star formation physics. At the conclusion of the project, important leaps have been achieved in our understanding of the initial phases of star formation.

On the theoretical front, self-consistent chemodynamical simulations in 1.5 and 2 dimensions have been developed, with and without magnetic fields, and with full integration of an updated chemical network. These simulations have been used to study a variety of effects and achieve important insights on various open problems:

i) The effect of chemical depletion on observations of the mass-to-magnetic-flux ratio has been theoretically investigated, and it was found that OH depletion can make the mass-to-flux ratio appear to decrease as one moves from lower to higher density regions, while in fact the opposite is the case. This is an extremely important result in interpreting observations of high- and low- density regions of the same cloud (more information is available at

ii) A new algorithm was developed for diagnosing the 3-dimensional shapes (oblate, prolate, or spherical) of molecular cloud cores based on chemical signatures of various easily observable molecules. More details on this method can be found at

iii) The physical origin of “striations” – hair-like structures that are observed in low-density clouds was identified. Using a systematic study (by means of a suite of simulations including extensive parameter studies) of all possible and proposed causes of this phenomenon, it was found that the only way to produce striations with properties that match the observed ones is the propagation of magnetogydrodynamic waves inside molecular clouds. More details on this work and the simulations that have made it possible can be found at

In addition, a a Python line radiative transfer code was developed that can be combined with the chemodynamical models developed for the SFOnset project and produce predictions of specific observables (molecular line profiles, line maps) that can be directly compared to real data. The paper describing the code is being prepared for submission to the Monthly Notices of the Royal Astronomical Society, and the code itself will eventually be freely available at the project website,

On the observational front, new data on interstellar clouds have been acquired and existing data have been re-analyzed to produce more robust tests of theories of the earliest stages of the star formation process. Such efforts include:

i) A re-evaluation of the ubiquitous claim that filamentary structures in molecular clouds, as recorded by the Herschel mission, appear to have a characteristic width. With more sophisticated and robust, unbiased statistical treatments, it was established that the existence of a characteristic width is not supported by current data. More information on this work can be found at

ii) A systematic study of the magnetic field, filamentary structures, and their relative orientation in the Polaris Flare interstellar cloud. The filamentary structures were probed using Herschel data, while the magnetic field was mapped using the polarization properties of background stars as measured by the RoboPol polarimeter, which is installed at the Skinakas Observatory. More information on this work, including all of the RoboPol stellar data in the Polaris Flare region can be found in and in

iii) A detailed study of appropriate observables to statistically study the filamentary structures that appear to thread molecular clouds. Strict criteria were defined for selecting structures that can be considered as physically coherent filaments. It was found that only a small fraction of “filaments” identified by topological analysis tools commonly used in the literature are true filaments in the physical sense. More information can be found in

Finally, in a new project that lies at the intersection of interstellar medium science, optical polarimetry, and cosmology, a new tomographic technique was developed to map in 3 dimensions the interstellar dust that emits polarized light and which constitutes the dominant foreground in searches for an inflationary signature in the polarization of the cosmic microwave background. SFOnset researchers will apply the new technique over large areas of the sky in the next years.

The theoretical predictions produced by the SFOnset project will become available on the project website,

Project SFOnset has been supported by the European Union’s Seventh Framework Programme, through a Career Integration Grant (CIG) Marie Curie Action, under grant agreement PCIG- GA-2011-293531.