My research interests lie at the intersection of data science and geoscience. I am interested in deriving scientific insights by utilizing the tools of data science to tackle geoscience problems, particularly as applied to remote-sensing of the cryosphere. This includes automated processing pipelines of remote-sensing data products, such as automated glacier calving front and grounding line delineation from a variety of satellite data sources. Furthermore, I’m interested in interpretable machine learning algorithms such as LRP in exploring complex dynamic systems.
We are at the cusp of a revolution in geoscience, and specifically glaciology, where the sheer volume of observational data will shift the current bottleneck in our understanding from data availability to processing ability and interpretability. Parallel advancements in computational tools utilizing big data and machine learning, cloud computing, and even advances in probabilistic programming and statistical inference, provide the perfect opportunity to push the limits of our understanding of changes in the cryosphere.
I’m also interested in improving our understanding of glaciological processes contributing to sea level rise. To that end, quantifying mass balance of glaciers at a regional level and understanding glacier dynamics are of crucial importance. I believe interpretable ML systems can also assist with an improved understanding of key, but difficult to model processes, such as marine ice cliff instability.
Lastly, I’m also interested in pursuing the social impacts of sea level rise. This includes remote sensing of coastal infrastructures as well as hierarchical Bayesian models of the multitude of variables influencing the social and economic cost of sea level rise.