Date of Award

8-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Physics and Astronomy

Committee Chair/Advisor

Dr. Dieter H. Hartmann

Committee Member

Dr. Marco Ajello

Committee Member

Dr. Mark Leising

Committee Member

Dr. Joan Marler

Abstract

Being the most energetic explosions in the Universe, Gamma-Ray Bursts (GRBs) can be detected back into the reionization era and help probe cosmic evolution. However, their use as cosmological tools is limited by unresolved aspects of their progenitor systems. Recent GRB detections demonstrate that the canonical prompt-duration classification is not a reliable method for distinguishing progenitors. This ambiguity also creates tension in the cosmic GRB rate density and its connection to the star-formation history. This tension stems from the unclear picture of the GRB formation efficiency. A central challenge is the GRB formation efficiency, dependent on the progenitor star metallicity ($Z$), host galaxy environment, an evolving initial mass function, supernova formation efficiency, binary evolution and merger delay times. It is degenerate with the luminosity evolution, selection biases and redshift completeness. This incredibly difficult and convoluted function of efficiency factor, proves to be a barrier in the GRB-progenitor connection. My dissertation addresses this complex puzzle from three complementary perspectives.   The first part of my thesis examines long GRBs (LGRBs) with X-ray plateaus, a common signature of GRBs with prolonged central engine activity. They are crucial for calibrating afterglow plateau correlations that separate progenitor channels and for training machine-learning methods to predict redshifts. Using a spectroscopic redshift sample, I infer its intrinsic rate density after the sample completeness and selection bias corrections. The results imply that the LGRB rate density (LGRB-RD) is shallower at high-$z$ and low-$z$ regimes relative to SFRD. The high-$z$ behavior is attributed to the possible $Z$ evolution or evolving IMF effect in massive star progenitors, and the low-$z$ nature is limited by statistical uncertainties and cross-channel contamination. A simple theoretical LGRB-RD model with an evolving power-law formation efficiency is insufficient to explain the observed rates, suggesting a need for a complex combination of physically motivated effects.    The second part addresses progenitor mixing more directly, due to cross-channel contamination and recent unconventional GRB detections, moving beyond canonical duration labels. I use a complete sample with spectroscopic redshifts and spectral information to derive an intrinsic GRB-RD corrected for observational selection biases. This rate density is modeled with a mixed population model containing the two most robust progenitor channels. The collapsar channel includes a composite function of the $Z$ bias and the residual high-$z$ evolution effect, and the merger channel includes a delay-time distribution. The inferred top-ranking models favor a hybrid channel over pure prompt or delay channels by integrating the rate density and luminosity information. The hybrid models exhibit prominent delayed contributions at low $ z$ with intermediate-to-long delays, and an increasing prompt-channel contribution toward the early Universe. Tests with different $\rm{T_{90}}$ cuts elucidate that many duration-selected samples require significant merger-like progenitor systems at $z< 2$, whereas the longer-duration cuts can include extreme delayed populations, such as WD/BH or WD/NS mergers, concentrated at $0

Author ORCID Identifier

0009-0002-2068-3411

Download

Available for download on Tuesday, August 31, 2027

Share

COinS