Lawrence Jing Yueh Liu

Welcome to my website! I am currently a PhD Student at Department of Climate, Meteorology, and Atmospheric Science (Climas) in University of Illinois Urbana-Champaign, and work with Prof. Kelvin Droegemeier. I am interested in understanding the weather system and improving its predictability with data assimilation and machine learning, especially in mesoscale regime.

I started my research career at National Central University, where studied data assimilation with Prof. Shu-Chih Yang. My master study is mainly focused on radar data assimilation. More specifically, the research is focused on how the environmental dynamic field adjustment can affect the validity of the precipitation system predictions. I used a multi-scale ensemble data assimilation system (WRF+LETKF) to study the cross scale interaction in the mesoscale precipitation systems. The research manifest the importance of larger scale environmental flow to the mesoscale system. The more accurate environmental flow can lead to better precipitation forecast significantly. This research also leaves several questions: 1. What is the optimal setting in localization length and the density of the datasets in this multiscale framework? 2. How can we add information in smaller scale (Scale-Dependent Inflation) 3. How to explain the major setback of not scale-awared background error covariance (BEC) structure?

Now I am exploring the best combination of data assimilation and machine learning techinques to make the efficient, accurate, and explainable mesoscale forecasts. Nevertheless, I’m still curious about the cross-scale interaction in the atmosphere. Is there a limit on the predictability of the nature? Is there a better explainable framework to make the statement of the question more clearly? Solving these questions would be interesting.

Facebook LinkedIn