Victor Canela and Jacob Ngaha from the University of Auckland are using NeSI’s Mahuika supercomputer to model and better understand the behaviour of photons. (Image by Gerd Altmann, Pixabay)

Understanding the behaviours of light

"It’s a new landscape we can explore with these techniques."

The Challenge: 

To build a real-world model of the quantum behaviour of light as it’s scattered from atoms.

The Solution: 

NeSI’s Mahuika supercomputer turning one big problem into many solvable smaller ones by running quantum trajectories through parallel processing. 

The Outcome: 

A better understanding of photon fluorescence, with applications in quantum information theory.


Humans encounter light during almost every moment of their waking day. More than any other sense, sight determines how we navigate the world. For this reason, light can seem almost mundane, but the truth is the opposite. The behaviour of photons – the fundamental unit of light – sits on the border between our classical understanding of reality and the counter-intuitive world of quantum physics. 

Dr Victor Canela and postgraduate student Jacob Ngaha are researching this phenomenon at the Department of Physics at the University of Auckland, alongside Prof Howard Carmichael. There, they are discovering more about the quantum nature of light by building an open system model of photon scattering with the help of NeSI supercomputing.

Photon scattering is the absorption and emission of photons by an atom, resulting in a  change of energy levels. Victor and Jacob’s research involves building a model of how photon scattering might look in a real-world setting. 

“Atoms emit photons when their energy drops to a lower level, and we can measure the fluorescence spectrum of these photons,” Victor explains. “We’re studying physical systems so we need to allow for environmental interaction. These systems are not perfect, so we need to account for flaws.”

While Victor and Jacob’s work is primarily focused on creating a better understanding of the quantum mechanics of light, with real-world applications in quantum information theory. Particular atomic structures are able to produce anti-bunched light, or beams of single photons of certain wavelengths or colours. This could be applied to quantum computer construction or communications projects, for instance China’s Quantum Science Satellite.

“It’s a way to produce fields which exhibit quantum characteristics. By studying them, physicists are able to learn more about the fundamental nature of light. Our end-goal is trying to understand these fluorescence spectra, though people are interested in these quantum fields for the purpose of encoding information.”

In order to understand how this fluorescence operates in a real experiment, Victor and Jacob need to run the model as a series of quantum trajectories. This involves breaking down the master equation for the model – which accounts for the complex time evolution of the atom-photon model – into a large number of smaller trajectories, which evolve randomly with time. These quantum trajectories can be solved with parallel processing, making NeSI participation vital in building the model. 

“We are running close to a hundred jobs at the same time and each job has separate parameters. It is difficult for us to run these jobs and keep track based on their parameters. NeSI was able to supply a script that keeps track of this automatically. They were also able to supply a speed increase by different methods,” says Victor. 

Victor and Jacob were able to run these quantum trajectories on NeSI’s Mahuika supercomputer. The NeSI computational science team also played a vital role in optimising the model’s calculations and the overall workflow.

“We worked with Chris [Scott] and Alex [Pletzer]. They were able to help greatly with how we work in the cluster and compile software. Alex is a veteran with Fortran. He points out lots of little things that we would never have figured out, to speed everything up.”

Chris and Alex worked closely with Victor and Jacob to profile their code and understand where most of the computational time was being spent. They were able to optimise the code as well as implement a more automated workflow on Mahuika. CMake — a suite of tools designed to build, test and package software — was introduced to make it easier to compile the code, experiment with different compilers, and run benchmark tests.

With the model created, Victor and Jacob are planning to publish their findings of how photon fluorescence reacts in laboratory systems. Victor says this field is open to a wide range of research possibilities, which he will be pursuing in the future.

“Not many people have looked at the way we’re looking at the light – including filtering and correlating output photons based on their different frequencies, or colours,” he says. “It’s a new landscape we can explore with these techniques – particularly using quantum trajectories. Computationally, you can do things with quantum trajectories that you wouldn’t be able to do any other way.”

This foundational work will support new quantum technology development, both in New Zealand and internationally. 

- - - - - -

Do you have an example of how NeSI platforms or expertise have supported your work? We’re always looking for projects to feature as a case study. Get in touch by emailing

Next Case Study

NeSI Case Study Erik Behrens Jonny Williams NIWA

Getting closer to more accurate climate predictions for New Zealand

"Researchers will be using this model for ground-breaking research."