Machine Learning, Climate Informatics, Causal Inference

Savini utilizes data science and applied machine learning to tackle scientific problems in Earth sciences. Her research has focused mainly on understanding causal interactions between climate processes (e.g., teleconnections) and their changes with anthropogenic climate change, as well as identifying sources of climate predictability in sub-seasonal to seasonal timescales. Before joining the ISWS in 2023, Savini completed a postdoctoral fellowship with the Department of Geophysics at the Colorado School of Mines, focusing on Machine learning applications in geological carbon dioxide sequestration, and spent a year at a climate tech startup researching on climate predictability. 

Ph.D. Computer Engineering, 2020
Colorado State University, Fort Collins, CO, USA 

BSc. Electronics and Telecommunication Engineering, 2014
University of Moratuwa, Sri Lanka

Samarasinghe, S.M., Connolly, C., Barnes, E. A., Ebert-Uphoff, I., and Sun, L. (2021),
Strengthened causal connections between the MJO and the North Atlantic with climate warming.
Geophysical Research Letters, https://doi.org/10.1029/2020GL091168. 

Samarasinghe, S.M., Deng, Y. and Ebert-Uphoff, I. (2019), A Causality-Based View of the
Interaction between Synoptic- and Planetary-Scale Atmospheric Disturbances. Journal of the
Atmospheric Sciences, https://doi.org/10.1175/JAS-D-18-0163.1

Ebert-Uphoff, I., Samarasinghe, S.M. and Barnes, E.A. (2019), Thoughtfully Using Artifi-
cial Intelligence in Earth Science. EOS – Earth and Space Science News, https://eos.org/opinions/thoughtfully-using-artificial-intelligence-in-earth-science. 

Barnes, E.A., Samarasinghe, S.M., Ebert-Uphoff, I. and Jason, F. (2019), Tropospheric and
Stratospheric Causal Pathways between the MJO and NAO. Journal of Geophysical Research:
Atmospheres, https://doi.org/10.1029/2019JD031024. 

Samarasinghe, S.M., McGraw, M.C., Barnes, E.A. and Ebert–Uphoff I. (2018), A Study of
Links between the Arctic and the Midlatitude Jet Stream Using Granger and Pearl Causality.
Environmetrics, https://doi.org/10.1002/env.2540.