As part of my research, I've developed various software for computational modeling of relativistic heavy-ion collisions and quark-gluon plasma. Here are a few highlights:


A new model for the initial conditions of heavy-ion collisions. The name is an homage to Trento, Italy, where my fellow Duke student Scott Moreland and I formulated the idea (it also stands for Reduced Thickness Event-by-event Nuclear Topology).

Visualization of an event generated by the model (click for interactive version on non-mobile devices):


Modern implementation of free streaming: a simple, common model for the brief period after the initial condition but before quark-gluon plasma formation (when hydrodynamics takes over).

Animation of free streaming a TRENTo event for 5 fm/c (approx. 10-23 seconds):


My published journal articles. See also my Inspire page.

Applying Bayesian parameter estimation to relativistic heavy-ion collisions: simultaneous characterization of the initial state and quark-gluon plasma medium

J. E. Bernhard, J.S. Moreland, S. A. Bass, J. Liu, U. Heinz

First Bayesian analysis using our initial condition model TRENTo. Presents a number of physical insights including the first systematic, quantitative estimate of the temperature-dependent shear viscosity of the quark-gluon plasma.

Quantifying properties of hot and dense QCD matter through systematic model-to-data comparison

J. E. Bernhard et. al.

My first paper applying Bayesian methods to characterize the quark-gluon plasma.

Talks & posters

I have presented my primary research project on Bayesian parameter estimation numerous times around the world (see my Speaker Deck for a complete list). These are the most recent:

Characterization of the initial state and QGP medium from a combined Bayesian analysis of LHC data at 2.76 and 5.02 TeV

My talk for Quark Matter 2017, the preeminent conference in heavy-ion physics.

Precision extraction of QGP properties with quantified uncertainties, part II: methodology and results

Overview of my research and recent results for a broad audience of statisticians and nuclear physicists. This was a shared talk; Steffen Bass presented the introduction.



photo of Jonah Bernhard
I study computational nuclear physics, focusing on modeling and characterizing the quark-gluon plasma (QGP), a fluid-like state of matter produced in tiny quantities by relativistic heavy-ion collisions. These QGP droplets exist for roughly 10-23 seconds—far too briefly to observe directly—so we rely on comparisons of model calculations to experimental data to infer properties of the QGP medium. I am applying statistical methodology such as Bayesian parameter estimation to rigorously quantify these properties. In addition, I have written several physics models central to the field's phenomenology and used them to simulate millions of heavy-ion collision events on the Open Science Grid and at NERSC.
I completed my PhD in Physics at Duke University in 2018. Previously, I graduated from Swarthmore College in 2011.
  • Computational and statistical modeling
  • data science
  • Bayesian parameter estimation and uncertainty quantification
  • Gaussian processes and machine learning
  • parallel and grid computing
  • Python
  • Cython
  • C/C++
  • Git and open-source software development
  • Linux/Unix shells and system administration