Christina Kumler

I have most recently been working with machine learning methods, included structures like neural networks and random forests, to better utilize data we obtain from inputs like numerical weather models and satellites. I use these models to solve different problems including region of interest identification and modeling fire power into the near future. My background in applied mathematics, atmospheric, and oceanic sciences lent itself nicely to this type of work and I hope that these tools can be used to help improve forecasts, better utilize the massive amounts of data we cannot process fast enough in "real-time," and more. 

Additionally, I am avid about outreach opportunities. These have included a range of audiences from fellow scientists to the public; examples include teaching machine learning tutorials to college and graduate students as well as speaking about machine learning and the weather to elementary and middle-schoolers.

The machine learning community in atmospheric science is growing rapidly, and I participate as the president of the Artificial Intelligence (AI) Committee president for the American Meteorological Society as well as in institutes such as AI2ES and CIRA. I lead the NOAA AI working group, which meets monthly, to discuss ongoing research, new advancements, etc. Basically, if you have AI in atmospheric sciences question, I can either help out or point you to someone who might be able to provide more information!