My name is Maya Maciel-Seidman and I am a Ph.D. student in Earth and Climate Sciences at Duke University's Nicholas School of the Environment, advised by Dr. Jonathan Ryan. I study Greenland Ice Sheet surface processes related to meltwater runoff using a combination of remote sensing, regional climate models, machine learning, and field observations. I am currently developing transfer learning models which are deep learning emulators of regional climate models, fine-tuned with in-situ observations, to predict catchment-scale meltwater runoff from the Greenland Ice Sheet.
Before arriving at Duke, I earned a B.A. in Earth and Environmental Sciences with a minor in Data Science from Vanderbilt University in 2025. At Vanderbilt, I developed new methodologies for quantifying residential carbon emissions as a researcher in the Climate, Health, and Energy Equity Lab. I previously served as a Naval Research Enterprise Internship Program (NREIP) intern at the U.S. Naval Research Laboratory's (NRL) Remote Sensing Division for two years. At NRL, I developed random forest algorithms to determine relationships between climate and active layer thickness and convolutional neural networks to delineate and map ice wedge polygons from UAV-based lidar-derived DEMs. As part of my research, I conducted fieldwork in Utqiagvik, Alaska.
When I’m not thinking about research, I spend my time rock climbing and kayaking, perfecting my challah recipe, trying to beat my NYT Crossword time (my fastest time is 2 min, 45 sec on a Monday), and cheering on the Vanderbilt Commodores.