Overall, my research area lies in the application of machine learning in geospatial datasets. I am working on a joint project between the departments of Geography and Computer Science and Engineering here at Michigan State University. The project is entitled “Robust Algorithms for Multi-Task Learning of Spatio-Temporal Data” funded by NSF in which I apply novel machine learning techniques for various geospatial-temporal problems. Specifically, I work with high resolution and large-scale spatiotemporal datasets mainly in the field of hydroclimatology. With the help of appropriate algorithms in machine learning, I examine whether we can obtain a more powerful modeling capability compared to more traditional techniques. For example, one of my recent projects was to investigate the feasibility of using machine learning methods to reproduce the US drought monitoring (USDM) maps generated by domain experts. The USDM is developed by multiple agencies to provide an accurate assessment of drought conditions in the US. The map is built using multiple physical indicators as well as reported observations of the local contributors. As the final product, the USDM experts use their best judgment to develop the drought map. My proposed framework basically tried to mimic the integrating process performed by the USDM experts by considering different input data scenarios and machine learning which produced very promising results.