Anna Krokhine ’24

Major

Mathematics

Project Title

Application of New Machine Learning Tools for the Study of Ooids and Their Ancient Tropical Marine Environments

My objective was to use applied mathematics and machine learning to analyze growth rings on ooids, a type of sedimentary grain. Ooids form their size and shape over time in response to various environmental factors such as ocean chemistry and current velocity. Observing their growth rings can provide key evidence toward understanding ancient ocean climate. I manually traced rings to create training data and built various programs to process and model these data. The most important tool we looked at were physics-informed neural networks (PINNs), which are neural networks enriched by partial differential geometry equations in our case. The idea was to use PINNs to write an algorithm that would parameterize ooid rings and indicate the factors behind their growth history. Throughout this process, I strengthened my image processing skills, got my first experience with machine learning, and learned how to incorporate work from the larger research community. I saw a fascinating application of mathematics (my concentration) to uncovering the Earth’s past climate, and I am looking forward to working more with machine learning.



Internship Year

2021

Project Category

Climate and Environmental Science

Organization(s)

Maloof Research Group, Department of Geosciences, Princeton University

Mentor(s)

Adam Maloof, Professor of Geosciences; Bolton Howes, Ph.D. candidate, Geosciences