Author ORCID Identifier
Degree Year
2024
Document Type
Thesis - Oberlin Community Only
Degree Name
Bachelor of Arts
Department
Physics and Astronomy
Advisor(s)
Dan Stinebring
Committee Member(s)
Jason Stalnaker
Rob Owen
Yumi Ijiri
Dan Styer
Dan Stinebring
Keywords
Pulsar timing, Gravitational wave detection, Optics, Stochastic modeling, Computer simulation, Interstellar medium
Abstract
Pulsar timing has shown significant potential in detecting gravitational waves in the last decade. Yet, current results are noisy, limiting the effectiveness of pulsar timing. It is well known that interstellar timing delays constitute one important source of these timing errors. However, the current methods meant to correct for interstellar timing delays significantly decrease sensitivity to low-frequency gravitational waves in pulsar timing arrays. In this thesis we show how current interstellar dispersion models can be biased by other timing effects including non-chromatic effects, like gravitational waves (GWs) themselves, and chromatic effects like scattering (§ 3.2.1 and § 4.2). We develop new methods to calculate unbiased estimates of chromatic timing effects to produce “chromatic-adjusted” timing residuals (§ 4.3), as well as introduce new methods that can be used to constrain the size of chromatic models to increase gravitational wave sensitivity (§ 3.3.1). Recent studies have shown limitations for current statistical theories on the timing delay produced by interstellar scattering, which, as a timing effect, has often been ignored in pulsar timing. Through simulations, we investigate path-length variations of radio waves in turbulent plasma, and show that the resulting stochastic behavior may explain these recent observational findings (§ 4.1). Specifically, we find that scattering timescales, τs, and observing radio frequencies, ν, are potentially best understood using a stochastic power law model τs ∝ να(t), compared to current models that assume a constant power law index of -4 to -4.4. This led us to search for potential scattering in the NANOGrav 15-year pulsar timing array using our new chromatic-adjustment method. Initial lower-bound analyses find that half of the pulsars in the PTA may be scattered (§ 5.3). Closer analysis of individual pulsars yields significant (p < 10−4 and much better) detections of scattering in the following pulsars: B1937+21, J1643-1224, J1600-3053, J1903+0327, J1744-1134, and J2145-0750 (§ 5.4). Our findings give evidence for time-correlated scattering, as well as significant bias from scattering in current dispersion solutions. In the case of J1903+0327, we find scattering on the order of 100 μs, strong enough that chromatic-adjusted residuals differ significantly from NANOGrav’s current best-fit model, suggesting a misfit model. Our analyses highlights the limitation of currently used chromatic models in pulsar timing.
Repository Citation
Weber, Felix E., "Detecting Gravitational Waves with Pulsars: Optimally Correcting for Interstellar Delays" (2024). Honors Papers. 910.
https://digitalcommons.oberlin.edu/honors/910
