Hope College Physics Department
Research Experiences for Undergraduates
Population Synthesis of Radio and Gamma-ray Millisecond Pulsars
|Student Name:||Sara Frederick|
|Student's Home Institution:||University of Rochester|
|Research Advisor:||Dr. Peter Gonthier|
|Source of Support:||
This work is made possible by the generous support of the National Science Foundation (Grant No. RUI: AST-1009731 and Grant No. REU: PHY/DMR-1004811) and the NASA Astrophysics Theory and Fundamental Program (NNX09AQ71G and 12-ATP12-0169) as well as internal funds from the Hope College Physics Department.
In recent years, the number of known millisecond pulsars (MSPs) in the Galactic disk has risen substantially due to confirmed detections by Fermi. This new population synthesis uses Markov Chain Monte Carlo techniques to explore the large and small worlds of the model parameter space and allows for comparisons of the simulated and detected MSP distributions. The simulation employs empirical radio and gamma-ray luminosity models that are dependent upon the pulsar period and period derivative with freely varying exponents. Parameters associated with the birth distributions are also free to vary. The computer code adjusts the magnitudes of the model luminosities to reproduce the number of MSPs detected by a group of ten radio surveys, thereby normalizing the simulation and predicting the MSP birth rates in the Galaxy. Computing many Markov chains leads to establishing preferred sets of model parameters that are further explored through two statistical methods. Marginalized plots define confidence regions in the model parameter space using maximum likelihood methods as well as Kuiper statistics from comparisons of cumulative distributions. These two techniques provide feedback as a consistency check to affirm the results. We follow an analogous set of assumptions that we have used in previous, more constrained Monte Carlo simulations. In addition, radio flux and dispersion measure constraints have been imposed on the simulated gamma-ray distributions in order to reproduce realistic detection conditions. The simulated and detected distributions agree well for both sets of radio and gamma-ray pulsar characteristics, as evidenced by our preliminary comparisons.
Publications and Presentations: