Please see (and apply at) the full job ad: https://lanl.jobs/search/jobdetails/machine-learning-and-hydrology-postdoctoral-researcher/949cb124-4144-40ba-872b-ae96323ecab7

We seek an intellectually curious postdoc interested in machine learning applications to hydrology. This position will involve research spanning multiple projects. One project will be aimed at applying long-short term memory (LSTM) networks or other forms of interpretable artificial intelligence to forecast river discharge in watersheds affected by wildland fire. Another project focuses on optimizing reservoir management using state-of-the-art hydrologic forecasting and deep reinforcement learning. Our team has developed a data platform that provides convenient access to large quantities of standardized hydrological data, allowing the researcher to focus more on modeling and less on data wrangling. For this position, machine learning expertise is essential, while knowledge of hydrology and wildland fire science is desirable but not required.

Some of your tasks will include:

• Building machine learning models (e.g. LSTMs, Transformers) for streamflow and flood predictionin basins affected by wildland fire and reservoirs
• Exploring techniques for building data-driven models that incorporate known physical principles
• Developing and deploying methods for understanding what your models learned

This position provides an opportunity to apply your skills toward a variety of impactful problems. Although your project will focus on modeling streamflow, domain-specific knowledge (e.g. hydrology, climate and/or earth sciences) is not required. Depending on your interests and time, you may have opportunities to contribute to other projects as well. Our teams will provide context, background, and guidance as you familiarize yourself with the domain-specific applications. While the overall research goals for these projects have been established, there is significant flexibility in the way these goals can be achieved, and novel approaches are encouraged.

​​​​​​​Please reach out to Chuck (cabolt@lanl.gov) with further questions about the opportunity.