Students,
A friendly reminder for CEE 500 seminar today at 3:30, which will be on zoom only.
Dr. Xiaodong Chen earned both MS and PhD degrees from our department. Xiaodong is currently a research scientist in the Atmospheric Sciences and Global Change division at the Pacific Northwest National Laboratory. He will talk about his career pathway and current research on predicting extreme hydroclimatic events using machine learning. Note that the seminar will be virtual only.
When/where: Thursday (11/10), on Zoom: http://depts.washington.edu/watersem/
Abstract: Improved understanding of the regional hydro-climate extremes from machine learning
The western U.S. hydro-climate system features significant footprints of both large-scale circulation patterns (atmospheric rivers) and local processes (precipitation and snowpack). These relevant extreme events thus exhibit multi-scale interactions with the large-scale hydro-climate environment, posing significant challenges to their predictability apart from their high impacts on the regional social-economic welfare. Over the past few years, machine learning (ML) has been proven as a powerful tool to explore and predict various aspects of the hydrologic cycle. In this talk, I will go over a few studies that involve different ML techniques and demonstrate their unique advantages as computationally efficient tools for non-linear hydrologic analysis of storms, snowmelt, and wildfires. They enable descriptions of hydrologic processes at various spatial-temporal scales and can effectively eliminate redundant information. I will also discuss how ML and big data (e.g., high-resolution hydroclimatic simulations) can benefit the development/evaluation of each other and how their synergy can elucidate new insights into the hydro-climate system at regional to global scales. Continue reading →