About Dr. Alexander Sun
Dr. Sun is a senior research scientist and computational hydrologist with an educational background in civil and environmental engineering. His areas of research include geostatistics, environmental modeling, risk assessment, and, more recently, data analytics. He has conducted research related to nuclear waste disposal, geologic carbon sequestration, groundwater and surface water modeling, and remote sensing.
Dr. Sun graduated from the University of California at Los Angeles in 1995 with a B.S. in civil and environmental engineering; he subsequently earned his M.S. and Ph.D. in civil and environmental engineering from the University of California at Berkeley, in 1996 and 2000 respectively. He was a graduate research assistant at Los Alamos National Laboratory for one year before going on to work as an engineer for Tetra Tech in Lafayette, California, and the Southwest Research Institute in San Antonio, Texas.
He has been with the Bureau of Economic Geology since 2011.
What are your current research activities?
I have a strong passion for process-based research and data-driven predictive analytics. My overall research objective is to apply analytical approaches to advance our current understanding of the impacts of climate change, human activities, and energy exploration on water resources, and to enable sustainable water- and environmental-management practices.
Currently, I am working on applying observational, experimental, numerical, and machine-learning tools to understand the water-energy-ecosystem nexus at multiple scales. I am also working on developing predictive models by integrating process-level modeling, deep learning, and real-time data.
What excites you the most about your current research?
I am most excited about the multidisciplinary aspects of my current research. I am excited to have opportunities to solve challenging environmental problems using state-of-the-art research tools and through collaboration with top-tier researchers both inside and outside the Bureau.
What is the desired outcome of your research?
I endeavor to conduct research that leads to tangible socioeconomic outcomes, facilitates environmental decision making, and protects public safety. I hope to accomplish these goals by developing models and decision support systems that help quantify and mitigate environmental risks, and by harnessing information from multisource earth-observation data in a timely and efficient manner.
What do you need in order to make your research efforts more successful?
A respectful work environment that cultivates creative thinking and fosters collaboration among groups.
What are your latest papers/publications, and what is most exciting about them?
In the past several years, I have devoted significant effort to adapting and demonstrating the merits of data science and machine-learning tools for solving hydrogeoscience problems. Examples of my recent publications include:
Sun, A. Y. and Scanlon, B. R., 2019, How can Big Data and machine learning benefit environment and water management: a survey of methods, applications, and future directions: Environmental Research Letters, v. 14, 073001, 28 p., doi:10.1088/1748-9326/ab1b7d.
Sun, A. Y., Scanlon, B. R., Zhang, Z., Walling, D., Bhanja, S., Mukherjee, A., Zhong, Z., 2019, Combining physically-based modeling and deep earning for fusing GRACE satellite data: can we learn from mismatch?: Water Resources Research, v. 55, no. 2, p. 1179–1195, doi:10.1029/2018WR023333.
Sun, A. Y., Zhong, Z., Jeong, H., Yang, Q., 2019, Building complex event processing capability for intelligent environmental monitoring: Environmental Modelling & Software, v. 116, p. 1–6, doi:10.1016/j.envsoft.2019.02.015.
Zhong, Z., Sun, A. Y., Jeong, H., 2019, Predicting CO2 plume migration in heterogeneous formations using conditional deep convolutional generative adversarial network: Water Resources Research, v. 55, no. 7, p. 5830–5851, doi:10.1029/2018WR024592.
Sun, A. Y., 2018, Discovering state-parameter mappings in subsurface models using generative adversarial networks: Geophysical Research Letters, v. 45, no. 20, p. 11,137–11,146, doi:10.1029/2018GL080404.
Who will benefit from your latest paper or publication?
A long-held view on machine-learning models is that they are black-box models without physical meaning or the ability to deliver mechanistic explanations of underlying physical processes. Because of this, the scientific community was and still is reluctant to accept black-box models. In our recent work, we demonstrate the merits of combining physics-based modeling and data analytics. We elucidate different ways for the geoscience community to benefit from the advent of deep-learning tools.
What was your most exciting past paper or publication, and why?
I have had the opportunity to collaborate with many top researchers over the years, but the book I coauthored with my father in 2015 has special meaning to me, both professionally and personally. Model calibration and parameter estimation serve as basic tools in environmental modeling. This book, designed for novel and experienced modelers, provides a systematic introduction on this topic and includes many examples.
Sun, N. -Z., and Sun, A. Y., 2015, Model Calibration and Parameter Estimation: For Environmental and Water Resource Systems: New York, Springer-Verlag, 611 p. doi:10.1007/978-1-4939-2323-6.
Who are the types of research partners you are seeking, and what skills or expertise could benefit your research?
A lot of my research is concerned with helping clients identify optimal management policies. Thus, I can benefit greatly from research partners who have such interests and needs and who want to leverage computational tools and historical data to come up with such strategies.
What are the desired relationships, expertise, or skills that could be brought in to benefit your research?
Models and decision support systems are pointless without potential users, stakeholders, and policymakers. Thus, relationships with government agencies and industrial partners are critically important for providing guidance on research directions.
What have been recent successes associated with your research?
Collaborating with Bridget Scanlon and UT’s Center for Space Research, we proposed and were recently awarded a 4-year NASA grant to study the feasibility of using data from GRACE (Gravity Recovery and Climate Experiment) and its follow-on mission, GRACE-FO, to monitor regional flooding potential. A major aspect of the study is related to data gap filling using machine learning.
This year I’ll also be leading two industrial collaboration projects that aim to develop real-time anomaly detection capabilities for different types of sensors using machine-learning algorithms.
What is the geographic location of your research?
The Lower Rio Grande, Central Texas, and other global river basins.