Applications of Scientific Machine Learning (SciML) into ecological time series data
Scientific Machine Learning (SciML) is a collection of novel Machine Learning algorithms that incorporate domain-specific information into their training process to get better forecasts and unveil novel information of the dynamics of the data.

I’m studying the capabilities of SciML algorithms to improve our forecasting and understanding of ecological time series data. I am particuarly interested in studying Neural Ordinary Differential Equations (NODEs) and its extensions to improve forecasting of ecological time series to inform management decisions.

Analysis of restoration strategies in bull kelp forest in California
Kelp forests are one of the most productive and biodiverse marine systems in the world. Rising temperatures in the ocean have increased kelp mortality, and restoration efforts have been rising worldwide to mitigate their impacts.

I use mathematical models to analyze the effectiveness of possible restoration strategies in the northen coast of California, where bull kelp coverage has decreased over 90% in a decade. These models are producing information useful for managers.