Duncan maps the Universe from the Big Bang to the Milky Way
"There is something really compelling to me about cosmology in general, that we are able to answer philosophical questions quantitatively" - Duncan Watts.
Duncan Watts came from the United States of America to begin working as a postdoctoral fellow at the Institute of Theoretical Astrophysics (ITA) in August of 2020, delayed a few months due to the global pandemic. After a bachelor’s and master’s degrees at Harvard College, in 2018 Duncan conferred his doctoral degree at Johns Hopkins University.
– For my PhD and for two years as a postdoc, I worked with the Cosmology Large Angular Scale Surveyor (CLASS) project.
Working with CLASS
CLASS is one of the many of astrophysical experiments trying to measure primordial gravitational waves from the Cosmic Microwave Background (CMB), the oldest light in our Universe imprinted on the sky when the Universe was just 380000 years old.
– My PhD project was essentially to show that what our telescope was proposing was actually possible: measure the tensor-to-scalar at high precision below current constraints, and with the predicted noise properties of the instrument, he explains.
During his doctorate, Duncan contributed in showing that polarized foregrounds, consisting of synchrotron radiation and thermal dust, can be removed from the cosmological background signal using CLASS data alone.
– In the group, I was essentially the only member doing theoretical work!
Until his first postdoctoral fellowship came:
– As a postdoc, I worked mainly on the analysis of preliminary CLASS data and characterizing its noise properties, assessing their consistency with external datasets, including the WMAP and Planck experiments. Notably, this was the first time in my scientific career that I worked anything close to raw data!
A "coherent" map of the Universe
– Basically, Cosmoglobe is our vision for the future of CMB data analysis. In the past 20 years, we have been moving more towards a model where people publish various chunks of their data; some experiments share maps of the sky, some experiments share their power spectra, and some experiments even share their raw time-ordered data.
Part of the Cosmoglobe project’s goal is to create a sense of coherence between all available datasets.
– All experiments should be observing the same sky, so when they disagree, it is hard to say what that means! Is one experiment just less sensitive than the other, is there some unmodeled systematic error in the data, or was the sky truly different when the two experiments made their observations? Prior to Cosmoglobe, there wasn’t a clear answer to this question, and there was no agreed-upon way to decide who was right, concludes Duncan.
Too much, too little
Cosmoglobe aims to create a framework where data from different experiments can be analyzed together, so that together they can be used to make maps of the sky.
– Through this, we are creating a model of the sky that uses all available data, and can be used by all scientists, making it easy for researchers to use the results of all previous CMB data.
– How will you accomplish that?
– The main engine behind Cosmoglobe is the Commander pipeline.
– A large part of this project involves writing new code and making the WMAP data accessible. For example, the raw WMAP data is almost 700 gigabyte, but with modern data compression techniques, we are able to make this raw data available with less than 200 gigabyte of required hard disk space. One of my goals is to make it so that any astronomer will be able to reproduce my analysis, he explains.
– What challenges do you encounter?
– I think the biggest challenge in this research is data; we both have too much of it and too little of it!
– At some level, we are trying to model the entire microwave sky, and we don’t have quite enough frequencies to constrain the individual components with high precision. At the same time, we are trying to analyze all of the data at the same time, which requires being able to process terabytes of data in a short period of time, adds Duncan.
– Much of our work now is focused on getting the best hardware and the most advanced computational techniques and squeezing them to their absolute limits, just so we can hold 2 terabytes in memory all at once and analyze it in less than a day, he says, excited.
Cosmology as the quantitative answer to philosophical questions
– What fascinates you about your research?
– There is something really compelling to me about cosmology in general, that we are able to answer philosophical questions quantitatively, begins Duncan.
– At the beginning of the 20th century, it was not a foregone conclusion that we would be able to say anything about the beginning of the Universe. Now, we are talking about observing the epoch of inflation, which would give us physical relics of the first 10-34 seconds of the universe. The fact that this is a question that we can meaningfully ask and even expect an answer still takes me aback once in a while, he concludes.
Norway: a nation with beautiful landscapes and introverts
Duncan came all the way from USA to Norway because of the CMB&CO research group.
– Back at the Johns Hopkins University I was aware of the Commander approach to CMB analysis. Once I received the offer, it was hard for me to think of a reason not to accept, since I had heard so much about the beautiful landscapes, the humane working conditions, and a nation full of introverts really struck a chord with me, he confesses.
– What is your experience at ITA so far?
– It was very interesting starting work during a pandemic, but everyone has been very supportive, and I’ve seen a bit of Norwegian culture seep through the zoom meetings.
Duncan enjoys living in Oslo, a really special city where nature is easily accessible.
– During the summer, I would be able to finish work, go to the gym, then take the train to Sognsvann to go for a swim before heading home and making dinner.
– Any expectations?
– I hope that I will continue to grow as a person and a scientist here. The colleagues I’ve met here have been all incredibly impressive, and I already have learned a lot just from collaborating with them. I also hope to take some much-needed time for rest and relaxation, either by swimming in the fjord or hiking in the woods.