Update: I will be starting as an Assistant Professor in the Atmospheric Sciences Department and the Wilkes Center for Climate Science & Policy at the University of Utah in January 2026. My research group will use ML and data-driven modeling to tackle problems at the nexus of air quality, fires, and climate extremes.

I am currently a NOAA Climate & Global Change postdoctoral fellow working with Noah Diffenbaugh in the Climate and Earth System Dynamics Group and with Marshall Burke in the Environmental Change and Human Outcomes Lab at Stanford University. My current research focuses on the effectiveness of prescribed fires as a wildfire mitigation strategy. The objective is to advance the mechanistic understanding of prescribed burning efficacy in a warming climate and to develop strategies to reduce the socio-ecological impacts of wildfires.

My PhD research with Daniel Jacob and Loretta Mickley at Harvard University’s Atmospheric Chemistry Modeling Group focused on (1) using machine learning to expand the capabilities of atmospheric chemistry models, (2) developing dimensionality reduction algorithms that can determine the optimal and equitable placement of air quality sensors, (3) investigating the potential for prescribed fires to abate wildfire smoke exposures in the Western United States, and (4) quantifying impacts of chemical data assimilation on air pollution forecasts for NASA’s GEOS Composition Forecasting model (GEOS-CF).

In the past, I worked as a junior research scientist at the University of Washington on air quality case studies and applications of machine learning methods to chemical mechanisms. I earned my B.A. in chemistry from Reed College, with research experience pertaining to air pollution monitoring and secondary organic aerosol modeling. In my spare time, I play jazz trombone, enjoy football/basketball, and watch horror movies.

Curriculum Vitae (last updated December 2024)

Contact information:

• mkelp@stanford.edu