A 2DVAR Blending Method for CYGNSS Wind Speed Observations

Xiaochun Wang, University of California Los Angeles, Los Angeles, CA, United States, Zhijin Li, JPL, Pasadena, CA, United States, Yuchan Yi, The Ohio State University, Division of Geodetic Science, School of Earth Sciences, Columbus, OH, United States, C.K. Shum, The Ohio State University, Division of Geodetic Science, School of Earth Sciences, Columbus, United States and Joel T Johnson, Ohio State University Main Campus, Department of Electrical and Computer Enginneering, Columbus, United States
Abstract:
A 2DVAR Blending Method for CYGNSS Wind Speed Observations

Xiaochun Wang, Zhijin Li, Yuchan Yi, C. K. Shum, Joel Johnson

The Cyclone Global Navigation Satellite System (CYGNSS) mission, launched in December 2016, provides ocean surface wind speed measurements from an eight satellite constellation using GPS reflectometry observations. The wind speed observations obtained provides unprecedented spatial and temporal coverage of the tropics, and are available even in regions of high precipitation. In this research, a two-dimensional variational method (2DVAR) is tested to map the along track CYGNSS wind speed observations and other satellite wind observations onto a regular spatial grid. Due to the irregular but frequent nature of CYGNSS samples, the creation of surface wind products for science using CYGNSS data requires careful consideration in order to achieve the full benefit of CYGNSS measurements. A major challenge in using 2DVAR is that CYGNSS provides wind speed and not direction measurements, while the relationship between wind components and wind speed is nonlinear. To address this limitation, we explore the combination of wind direction information from the background field and dynamical constraints to optimally determine wind directions along with wind speeds. Results for the Indian Ocean region will be presented.

Back to: Data Science for Modern Oceanography: Statistics, Machine Learning, Visualization, and More IV Posters