Skip to content
 

Marie McGraw

Research Scientist I

About Me:

she / her

Research Scientist I

marie.mcgraw@colostate.edu


Cooperative Institute for Research in the Atmosphere
Colorado State University
1375 Campus Delivery
Fort Collins, CO 80523-1375

CIRA 39

About Me: I am a research scientist at CIRA in the Machine Learning and Tropical Cyclone research groups. I love using the power of statistics, data science, and machine learning to understand a variety of weather and climate data. At CIRA, my research has been centered around using machine learning to understand tropical cyclone behavior and improve tropical cyclone forecasts. Some of my recent projects include:

  • Using AI-generated passive microwave imagery to study tropical cyclone structure and evolution [AMS 2025 talk];
  • Using machine learning to better quantify uncertainty of tropical cyclone intensity and track forecasts [check out some of our forecast products at tc-realtime];
  • Evaluating uncertainty quantification with neural networks for earth science applications [paper led by Kathy Haynes of CIRA];
  • Improving tropical cyclone forecasting with causal feature selection methods [AMS 2024 talk by Prof. Tom Beucler];
  • Categorizing bias in machine learning for earth science [paper led by Amy McGovern of the University of Oklahoma];

I’ve also given some general audience talks about machine learning and other computational approaches for weather and climate science:

While my current research at CIRA has been focused on tropical cyclones, I have worked on a wide variety of topics in weather and climate science. During my Ph.D. at Colorado State University, I studied extratropical climate dynamics using causality frameworks (1, 2, 3). Before I joined CIRA, I was a postdoc at the University of Washington where I analyzed forecasts of extreme sea ice loss events. I have also been involved in studies about jet stream dynamics, atmospheric moisture transport, and the dynamics of the intertropical convergence zone.