Stefan Osłowski


My research focuses on neutron stars and high time-resolution radio astronomy. They are extremely versatile laboratories of physics and a plethora of physical experiments is available when studying pulsars. They are the only objects for which all four fundamental forces play an important role

UTMOST is a Swinburne led experiment using a refurbished Molonglo Synthesis Telescope to search for the mysterious Fast Radio Bursts (FRBs). I'm also coordinating Swinburne's involvement in the CRAFT project, a similar FRB search campaign but running on ASKAP, a Square Kilometre Array pathfinder. Currently there are more theories about their origin than known events. We are hoping to understand their nature by significantly increasing the number of known FRBs as well as providing first accurate localization.

During my PhD candiduature, my main focus was on trying to push the limits of pulsar timing. Many pulsars can be timed to hundreds of nanoseconds. This is a very impressive result, yet still below the theoretically attainable precision. One striking example is the millisecond pulsar J0437-4715, observed regularly as part of the Parkes Pulsar Timing Array. During my PhD my supervisors and I tried to understand why our measurements are not as precise as we expect. This included identifying and correcting the systematic effects in pulsar timing and improving our theoretical understanding and predictions of pulsar timing precision.

Other areas that I have either worked on or am working now include: Araucaria project, population synthesis of neutron stars, semi-analytic models of galaxy formation, properties of black widow pulsars, and development of LOFAR's pulsar observing capabilities in the single station mode.

Image credit:NASA

Pulsar timing

Pulsar timing is an important methodology of astrophysics with a plethora of applications in many areas, most prominently in detection of gravitational waves with pulsar timing array experiments. Other applications include, but are not limited to:

Relativistic gravity.

Many pulsars are in very relativistic binaries enabling stringent tests of GR and alternative theories of relativity. Nobel prize winning observations of one of the binary pulsars provide evidence of existence of gravitational waves.


First exosolar planets where discovered with pulsar timing and the most precise published mass measurements of the Jupiter is also based on pulsar timing.

Interstellar medium

We can study the distribution of electrons in the Galaxy thanks to the dispersion of pulsar's signal.

Matter in extreme conditions
The equation of state of dense matter is constrained by the masses of the neutron stars - extremely dense objects. Pulsars are also perfect laboratories to study highly magnetised plasmas.


Majority of projects and work described here would not have been possible without help from numerous people. The main players are Michael Kramer, Joris Verbiest, Olaf Wucknitz, Andreas Horneffer, James Anderson and many others.


While working at the Bielefeld University and Max Planck Institute for Radioastronomy I am responsible for undertaking observations with most of GLOW LOFAR stations. Occassionally I use also the Swedish station maintained by the Onsala Space Observatory. Every week we observe about 115 pulsars for a total of 12 to 15 days per week (3 days per station on multiple stations). You can find more information on a separate webpage I maintain here. The large amount of observing time and high cadence allows to undertake studies which might be difficult otherwise.

Monitoring the interstellar weather

I am part of the project that studies the dispersion measure variations towards many of our 115 target sources. Some results are available in Julian Donner's thesis, a former Bachelor student in the radio astronomy group in Bielefeld. I am working on two publications related to this topic. The first one adresses the short-term variability of the dispersion measure towards PSR J1509+5531, a nearby classical pulsar travelling through the interstellar medium at very high velocity. The other project deals with the chromaticity of dispersion measure and uses both LOFAR and Ultra BroadBand backend on the Effelsberg 100-metre Radiotelescope.

The Bielefeld radio astronomy group is also involved in a project studying the scintillation of pulsars. First results for PSR J0814+7429 are available in a bachelor thesis by Hauke Jung. We are now working on a publication showcasing LOFAR single-station capabilities in an upcoming publication with Dan Stinebring and his student Jason Rosenblum who visited us for a few months in 2014. We undertake weekly observations of PSR J1136+1551 nearly simultaneously with Arecibo observations. This project is in collaboration with Dan Stinebring, Jim Cordes, and others.

Pulsar moding

Another project undertaken with LOFAR data pertains to the topic of pulsar moding. We have studies moding properties of a few pulsars, including B0329+54. The project has led to one bachelor thesis by Mario Lachetta and a master thesis by Alexandros Filothodors (University of Zielona Gora, Poland, not available online).

Technical work, student training, and sharing expertise

Large fraction of my time in the years 2013 and 2014 went towards improving the observing system for LOFAR stations in Germany as well as a visit in Sweden in May 2014. The latter resulted in a fully working pulsar backend there which we now use on a regular basis.

Poland will soon (hopefully before the end of 2015) host three LOFAR single stations. We have been involved in helping the member of POLFAR collaboration with technical aspects of observing such as choice of hardware and training students to use the pulsar backend. We hosted two students from Zielona Gora in 2014, which was partially funded through my Humbold Research Fellowship funds. I will travel to Poland in late 2015 to install and test the pulsar backend for use with Polish LOFAR stations.

Research highlights

Limitations to pulsar timing

During my PhD I've worked on PSR J0437-4715, the brightest nearby millisecond pulsar lying off the Galactic plane. This is pulsar provides a unique opportunity to study pulsar properties in unprecedented detail. I present a thorough analysis of this pulsar in the context of timing limitations in Osłowski et al. 2011. We have discovered that the timing precision and accuracy of this pulsar is already limited by the presence of noise intrinsic to the pulsar. This is a very important discovery as this noise will be a dominant factor for future timing experiments on the Square Kilometer Array. For current generation of telescopes only the brightest pulsars are affected by it, but we need to study which ones.

Properties of this noise were unknown until my publication. I demonstrated this noise heteroscedastic (i.e., has different variance in every phase bin) and temporally and spectrally correlated. Especially the latter property was unexpected and has severe implications for broadband receivers. It implies that we wont' be able to get as much improvement from wider bandwidths as we would expect purely from the radiometer equation. In this paper as well as its follow up Osłowski et al. 2013 we demonstrated that the deleterious effects of this extra noise can be partially mitigated. I encourage you to contact me or read the paper to learn more about this.

Owing to the technological and algorithmic developments, we were able to present a study of polarimetric properties of this pulsar in yet another publication, Osłowski et al. 2014. Along with presenting polarimetric properties we attempted to improve the pulsar timing precision by selectively integrating the pulse profile in hope of reducing the amount of pulsar-intrinsic noise.

A similar idea was tested in Shannon et al. 2014 in which I was strongly involved as well. However, the main focus of this paper is measuring the limitations of timing precision due to the aforementioned noise in a larger sample of millisecond pulsars, see Table 3 in this paper. In addition, we presented a better way of incorporating this additional noise into the estimates of time-of-arrival uncertainty. Traditionally this was done based on the timing model and our method is based on the pulse profiles allowing a better-motivated correction of the uncertainty estimate. For more details see Fig. 22 and relevant text.

Popluation synthesis

Krzysztof Belczyński created a stellar evolution and population synthesis code StarTrack. Building on it, we have created a synthetic population of double neutron star systems. Such systems are an important source of gravitational waves for ground-based interferometric detectors such as LIGO. In order to perform a detection, scientists analysing LIGO data need to have a waveform template of the expected signal. The waveform will depend on the properties of the emitting binary. In Osłowski et al. 2011 we investigate if the population of double neutron stars seen in radio has the same properties as the population of sources that will be seen in the gravitational wave window.

Software development

I'm an active user and/or developer of the following very useful pulsar packages:

In case you would like some features implemented in the above codes, especially first one of those, please contact me and I will try to help.

A somewhat common mistake is trying to install dspsr after installing the "stable" release of PSRCHIVE. Please ignore the stable release and always deploy the latest development version!

Keeping up with the astro-ph

Keeping up with the flood of papers on astro-ph / ADS can be difficult. Ben Barsdell wrote an excellent tool called astroPHilter that helps with this. After he got a job in industry he had no time to maintain it and I've taken over development and maintenance. I recommend you give it a go and see if you like it. If you have requests or suggestions I might implement them if I'll have spare time!

A related problem is tracking recent important papers and news. Journal clubs are a good way of keeping in touch with progress in areas different to your own. I've written a simple tool which provides suggestions for published articles and news from a few sources.

© 2016-2017 Stefan Osłowski