Akshit Deshpande | 2026 I.S. Symposium
Name: Akshit Deshpande
Title: Combining Visual and Quantitative Methods to Identify Disc Galaxies around Cosmic Noon in the COSMOS-Web Survey
Major: Physics
Minor: Statistical & Data Sciences
Advisor: Laura DeGroot
One of the biggest open questions in astronomy is how galaxies like our Milky Way formed and evolved over billions of years. To answer this, astronomers study galaxies as they appeared in the early Universe, but this comes with a challenge: the farther away a galaxy is, the smaller and harder it becomes to examine. This thesis focuses on galaxies around “cosmic noon,” a period roughly 10 billion years ago, when star formation across the Universe was at its peak. Using high-resolution images from the James Webb Space Telescope (JWST), this work develops a method to identify flat, rotating disc galaxies, like the Milky Way, among the many galaxies visible in deep-sky observations, a necessary step to study the method in which these galaxies grow and evolve with time. A sample of 50 galaxies was analysed using both visual inspection and computational tools that measure shape, structure, and light distribution. The results show that combining visual classification with quantitative measurements produces far more reliable identifications than either method alone. Certain shape-based measurements, particularly how concentrated and how asymmetric a galaxy’s light appears, proved most useful in distinguishing disc galaxies from rounder, elliptical ones. Other measurements, like galaxy smoothness, turned out to carry very little useful information at these distances. This work presents a tested, practical toolkit for classifying distant galaxies, highlighting that even with cutting-edge data, galaxy classification at cosmic noon remains a nuanced challenge that requires multiple lines of evidence.
Posted in Symposium 2026 on May 1, 2026.
