Postdoctoral Research in Machine Learning and Climate Modeling


Los Alamos National Laboratory is a multidisciplinary research institution engaged in science and engineering on behalf of national security. The climate modeling effort at LANL, in part, develops and uses advanced multi-scale models for the study of Earth system processes. This two-year position is available immediately.

The postdoc researcher will work in a Department of Energy (DOE) program that supports fundamental, interdisciplinary research to achieve an improved understanding of climate processes and to leverage such understanding to improve predictive skill of models. Exploiting the capabilities of DOE’s High Performance Computing (HPC) systems in order to accelerate advances in climate science is a priority.

Sophisticated, architecture-aware, variable-resolution dynamical cores developed under this project permit direct numerical studies of phenomena that are poorly resolved in the current generation of climate models. For example, in the atmosphere, such processes include effects due to steep topography, convection and cloud processes, gravity waves, frontogenesis, and others, and in the ocean they include interaction of submesoscale eddies and internal waves with larger scale circulation and topography. While applicants can propose their own research ideas, general research themes include (a) investigating and characterizing linkages between such processes and larger scale dynamics and biases, (b) characterizing and understanding model uncertainties, (c) developing alternative approaches to climate predictability ( and parameterizing the effects of unresolved dynamics and physics by leveraging recent developments in statistical and machine learning and model reduction techniques, etc.