Dr. Yixin ‘Berry’ Wen
Assistant Professor
she/her/hers
Focus Areas
Research Statement
My motivation to pursue more accurate remote sensing precipitation retrievals at global scale is powered by the massive precipitation observations and the advanced machine learning (ML) technologies. During my 10-year work on the advanced weather radar at NOAA and on various satellite missions (e.g. AIRS, GPM, MODIS etc) at NASA, I have been involved in development and evaluation of weather radar and satellite precipitation retrieval algorithms. My research strengths in remote sensing precipitation estimation and ML research include three research topics.
- 1) Validation of remote sensing products. The cross-validation among different remote sensing products or the ground validation of satellite and ground weather radar products are providing the essential error information associated with the products. When using ML models to ingest data from multiple sensors, the quality and errors associated with each datasets are important. High-quality training datasets are critical for most machine learning, especially deep learning algorithms. It is well known that the usefulness of these ML/DL algorithms highly depends on the quality of data used for ML/DL training. So, a key factor that limits ML/DL algorithms’ adoption in Earth science applications is the lack of highly accurate training datasets. My expertise on multiple data sources from both NASA and NOAA is a solid foundation for the future ML research.
- 2) Synergy of ground weather radar and satellite products. Remote sensing is the only means to monitor the distribution of precipitation on a large spatial scale. Both satellite and ground radar communities have made significant advances separately. It’s a golden era when massive remote sensing measurements meet ML technologies. In our recent study, we tested an ML/DL model pipeline on data from the Advanced Microwave Sounding Unit (AMSU) connected with the ground radar precipitation estimates as reference. The study provided insights in a successful ML-rain rate retrieval. Moreover, it also helped interpret the variables and impact factors of the “black box” ML algorithm, a common objection of scientists concerning ML. In the future study, I propose to overcome the ‘black box-syndrome’ challenge by creating rainfall algorithms for satellite data using ML with a careful variable impact analysis. A confidence flag product will be created from the variable impact analysis to give confidence and robustness about the accuracy of our retrievals.
- 3) Applications of remote sensing data to monitor and forecast natural hazards. I believe that the goal of our research is to benefit the society and quality of life. The application of remote sensing data to monitor and forecast natural hazards such as flash flooding, drought, tornado, etc.
Recent Funded Projects
NASA Jet Propulsion Laboratory, California Institute of Technology A19-0075-001, Ground validation for the Atmospheric Infrared Sounder (AIRS) retrieval system (PI, $139,000);
University of Oklahoma Cooperative Institute for Mesoscale Meteorological Studies (CIMMS) Director’s Directed Research Fund, Understanding satellite-derived cloud properties using polarimetric classification of hydrometeor types from WSR-88D radars. (PI, $20,000)
NASA Oklahoma EPSCoR Research Initiation Grant, Development of a joint ground polarimetric radar and satellite database. (PI, $27,000)
NOAA/NSSL, The impact of radar interference on WSR-88D CODE: The Common Operations and Development Environment
Recent Publications
M. Weber, Hondl, K., Yussouf, N., Jung, Y., Stratman, D., Putnam, B., Wang, X., Schuur, T., Kuster, C., Wen, Y., Sun, J., Keeler, J., Ying, Z., Cho, J., Kurdzo, J., Torres, S., Curtis, C., Schvartzman, D., Boettcher, J., Nai, F., Thomas, H., Zrnić, D., Ivić, I., Mirković, D., Fulton, C., Salazar, J., Zhang, G., Palmer, R., Yeary, M., Cooley, K., Istok, M., & Vincent, M., 2021, Towards the Next Generation Operational Meteorological Radar, Bulletin of the American Meteorological Society, 102(7), E1357-E1383.
Y. Wen, T. Schurr, C. Kuster and H. V. Vergara, 2021, Effect of Precipitation Sampling Error on Flash Flood Monitoring and Prediction: Anticipating Rapid-Update Weather Radars, Journal of Hydrometeorology, 22(7), 1913-1929..
Z. Li, Y. Wen, M. Schreier, A. Behrangi, Y. Hong, B. Lambrigtsen, 2020, Advancing Satellite Precipitation Retrievals with Data Driven Approaches: Is Black Box Model Explainable? Earth and Space Science, 8(2), e2020EA001423.
B. Coffer, M. Kubacki, Y. Wen, T. Zhang, C. Barajas, and M. K. Gobbert. Using Machine Learning Techniques for Supercell Tornado Prediction with Environmental Sounding Data. Technical Report HPCF-2020-18, UMBC High Performance Computing Facility, University of Maryland, Baltimore County, 2020.
B. Coffer, M. Kubacki, Y. Wen, T. Zhang, C. A. Barajas, and M. K. Gobbert. Machine Learning with Feature Importance Analysis for Tornado Prediction from Environmental Sounding Data. Proceedings in Applied Mathematics and Mechanics, 20(1), e202000112.
Z. Li, M. Chen, S. Gao, Z. Hong, G. Tang, Y. Wen, J.J. Gourley, Y. Hong, 2020, Cross-Examination of Similarity, Difference and Deficiency of Gauge, Radar and Satellite Precipitation Measuring Uncertainties for Extreme Events Using Conventional Metrics and Multiplicative Triple Collocation, Remote Sensing, 12(8), 1258.
Z. Li, G. Tang, Z. Hong, M. Chen, S. Gao, P. Kirstetter, J. Gourley, Y. Wen, T. Yami, S. Nabih, Y. Hong, 2020, Two-decades of GPM IMERG early and final run products intercomparison: Similarity and difference in climatology, rates, and extremes, Journal of Hydrology, 594, 125975.
Y. Wen, A. Behrangi, H. Chen and B. Lambrigtsen, 2018, How well were the early 2017 California Atmospheric River precipitation events captured by satellite products and ground-based radars?Quarterly Journal of the Royal Meteorological Society, 144, 344-359.
B. Lambrigtsen, H. V. Dang, J. Turk, S. Hristova, H. Su and Y. Wen, 2018, All-weather tropospheric 3-D wind from microwave sounders, IEEE J. Selected Topics in Applied Earth Observations and Remote Sensing, 11(6), 1949-1956.
Y. Gou, Y. Ma, H. Chen, and Y. Wen, 2018, Radar-derived Quantitative Precipitation Estimation in Complex Terrain over the Eastern Tibetan Plateau, Atmospheric Research, 203, 286-297.
Educational Background
- Ph. D. in Meteorology, University of Oklahoma, 2015
- M.S. in Geoinformatics, University of Oklahoma, 2012
- B.S. in Ecology, East China Normal University, 2006
Current Graduate Students
Currently accepting applications for graduate students to begin in Fall 2022.