Skip to main content Skip to secondary navigation

Climate System Predictability

Main content start
Schematic of the neural network architecture that incorporates internal variability and external forcing into a Combined Network. (From Gordon et al., in revision.)

 

Emily Gordon is a Data Science Postdoctoral Fellow with research interests at the intersection of climate variability and predictability, and novel data science methods. Her research is focused on developing machine learning approaches to understand the predictability of the near term impacts of climate change.

Climate system predictability is the study of predictable behavior in the Earth system across timescales. While the atmosphere is chaotic, the coupling between the ocean, land and ice systems leads to processes that evolve on longer (seasonal to decadal) timescales. Emily uses data-driven techniques to examine the predictability of this low frequency variability in the climate system. This involves investigating predictability across Earth system models and observations, and developing methods to best utilize the information available from troves of climate data. In particular, she is interested in using advances in AI to identify times and places where we can expect more predictable behavior in the system, and whether we can use this information to better understand, and prepare for, future climate change.

Related Publications & News