For my undergraduate degrees, I studied at Indiana University, where I received two B.S. degrees: one in mathematics and one in music performance with an outside field in computer science. During the summer after my junior year, I was fortunate to participate in the NASA Summer Institute on Atmospheric and Hydrospheric Sciences at NASA Goddard Space Flight Center. I loved the research project I worked on, which was focused on optimizing satellite CO2 retrievals. Following this, I decided to continue my studies and attended Colorado State University, where I received both a M.S. and Ph.D. in atmospheric science, focusing on the carbon cycle and land-atmosphere interactions.
Since I was a kid, I always loved to travel, and again I was fortunate to obtain a research scientist position at the Commonwealth Scientific and Industrial Research Organisation (CSIRO). While living in Melbourne, Australia, I assisted in developing their global climate model. After two years, I moved back to Fort Collins, CO, to continue work as a research scientist at CSU in the Department of Atmospheric Science. Continuing to focus on land-atmosphere interactions and the carbon cycle, my primary role was to update a complex land surface model, the Simple Biosphere Model (SiB). I designed, developed, and documented SiB4, which is a mechanistic, prognostic land surface model that integrates heterogeneous land cover, environmentally responsive prognostic phenology, dynamic carbon allocation, and cascading carbon pools. SiB4 is available on GitLab, and global hourly, daily, and monthly output for 2000-2018 is archived on the NASA ORNL DAAC. For more information on SiB4, please see:
While working at CSU, I took advantage of my employee benefits to obtain a Graduate Certificate in Business Intelligence in 2019. This rekindled my enjoyment with machine learning, and I decided to obtain a Masters Degree in Computer Science (M.C.S.) in 2020. For class projects, I enjoyed applying machine learning to environmental science tasks, and in 2021 I started working at CIRA with the hopes of combining my computer science skills with my atmospheric science background. Since being at CIRA, I have applied machine learning to a wide array of projects:
- Detecting low clouds
- Improving vertical profiles of temperature and moisture
- Automatic detection of gravity waves
- Creating synthetic microwave imagery
- Developing a new ensemble global genesis index for tropical cyclones
As a researcher who understands the importance of reliable datasets, ranging from in situ measurements to tower-based eddy covariance data to aircraft measurements to satellite data, I have recently been involved in a project aimed at developing AI-ready data standards, tools, and benchmarks. Using the TC PRIMED dataset, I have been helping to assess this dataset while providing feedback to the NOAA Center for AI (NCAI) data readiness sub-team on the AI-ready data guidelines. I am also helping to develop Jupyter notebooks providing training material on how to apply TC PRIMED to machine-learning applications.
I am also a collaborator with the NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography (AI2ES). In working towards developing trustworthy AI for weather and climate, I have recently been enjoying working on incorporating skillful uncertainty estimates into machine learning predictions. A team of us here at CIRA recently published a paper on how to create and evaluate uncertainty estimates with neural networks, which is accompanied by a series of Jupyter notebooks demonstrating the methods. I am also working with a risk communication team on how to effectively visualize and communicate uncertainty estimates, particularly for images.